LeaderboardBack to leaderboard — loading return destination
Opening report…
Loading report overview
M
MCP server
server.example.com/mcp
Tools and resources for your MCP client, with a public quality report and detailed checks.
Protocol
MCP 2025-11-25
Server version
1.0.0
Website
server.example.com
Supports
Tools · Resources
Overall score
90 /100
How is this score calculated?
Server metadata
Loading server metadata
Reviewed90/100
Discoverability
Loading discoverability
Reviewed90/100
Protocol
Loading protocol
Reviewed90/100
Tools
Loading tools
Reviewed90/100
MCP Apps
Loading mcp apps
Reviewed90/100
Tool documentation quality
Loading tool documentation quality
Reviewed90/100
LeaderboardBack to leaderboard — loading return destination
T
TimeToPost
api.timetopost.co/mcp
Plan, write, schedule and publish social posts across 25+ networks from chat: drafts with approval, best posting times, engagement stats, short videos, SEO blog articles and campaign links, all in your TimeToPost workspace.
Plan, write, schedule and publish social posts across 25+ networks from chat: drafts with approval, best posting times, engagement stats, short videos, SEO blog articles and campaign links, all in your TimeToPost workspace.
Writing style for any post content you draft: write like the account's human owner, not like an AI. NEVER use em dashes, en dashes or double hyphens; use a period, comma or colon instead. Skip hashtag spam (0-1 hashtags), filler openers ("Exciting news!") and rocket/sparkle emoji. Before drafting, read the account's recent posts with list_posts and match their voice, casing and rhythm. For X specifically: keep a calm, factual, receipts-not-rhetoric tone; X ranking buries combative or bait-y posts even when engagement is high, and a mute or block outweighs hundreds of likes. Never write engagement-bait. Use 0-1 hashtags on X (more than 2 trips spam detection). Structure threads as one hook tweet, then 5-7 self-contained value tweets, then a single CTA. Prefer native vertical video (9:16, roughly 45-90 seconds) when media is available; it substantially outreaches text. Never draft identical or near-identical content for two different accounts. These heuristics were verified July 2026 and decay; treat as current-best, not gospel.
Discoverability
2/6 passed2 of 6 passed33/100
Client listings and public setup information.
mcp.json or server-card.json
Either mcp.json or an MCP Server Card declaring this endpoint passes.
The MCP Server Card declares this server's exact endpoint.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description92/100
CAPABILITY: `read`. Discover static TimeToPost use-case guides by id, category, or case-insensitive words in task and primary keyword. This read-only operation performs no mutation, network request, paid-provider call, or automatic execution. An id returns the full guide, exact MCP prompt, actual hosted tool names, and status. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs75/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
query
Optional words matched case-insensitively against task and primary keyword.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
CAPABILITY: `read`. Build campaign tracking URLs from supplied URL and UTM records. This read-only, deterministic operation performs no mutation or network request. Returns a JSON array of tagged URLs, one per input record in the same order, or {error: "invalid_input"} when a record is malformed. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs89/100
Remove repeated schema validation from descriptions; focus on meaning and usage context.
records › items › url
Destination page as a full http(s) URL. Existing query parameters are kept; any existing utm_* values are overwritten.
Clear enough to supply this argument.
records › items › source
Where the traffic comes from (utm_source), e.g. newsletter or linkedin.
Clear enough to supply this argument.
records › items › medium
Channel type (utm_medium), e.g. email, social or cpc.
Clear enough to supply this argument.
records › items › campaign
Optional explicit campaign. Defaults to <brand>-<UTC year>-<UTC month> when brand is supplied.
Clear enough to supply this argument.
records › items › brand
Brand or niche slug used to name the campaign when campaign is omitted.
Clear enough to supply this argument.
records › items › term
Optional paid keyword (utm_term).
Clear enough to supply this argument.
records › items › content
Optional variant label (utm_content) to tell links in the same campaign apart.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "build_campaign_urls",
"description": "CAPABILITY: `read`. Build campaign tracking URLs from supplied URL and UTM records. This read-only, deterministic operation performs no mutation or network request. Returns a JSON array of tagged URLs, one per input record in the same order, or {error: \"invalid_input\"} when a record is malformed. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"records"
],
"properties": {
"records": {
"type": "array",
"items": {
"type": "object",
"required": [
"url",
"source",
"medium"
],
"properties": {
"url": {
"type": "string",
"minLength": 1,
"description": "Destination page as a full http(s) URL. Existing query parameters are kept; any existing utm_* values are overwritten."
},
"term": {
"type": "string",
"description": "Optional paid keyword (utm_term)."
},
"brand": {
"type": "string",
"description": "Brand or niche slug used to name the campaign when campaign is omitted."
},
"medium": {
"type": "string",
"minLength": 1,
"description": "Channel type (utm_medium), e.g. email, social or cpc."
},
"source": {
"type": "string",
"minLength": 1,
"description": "Where the traffic comes from (utm_source), e.g. newsletter or linkedin."
},
"content": {
"type": "string",
"description": "Optional variant label (utm_content) to tell links in the same campaign apart."
},
"campaign": {
"type": "string",
"minLength": 1,
"description": "Optional explicit campaign. Defaults to <brand>-<UTC year>-<UTC month> when brand is supplied."
}
}
},
"maxItems": 20,
"description": "One entry per link to tag; the output array matches this order."
}
}
},
"annotations": {
"title": "Build Campaign URLs",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
CAPABILITY: `read`. Calculate engagement rates for supplied post metrics. This read-only, deterministic operation performs no mutation or network request. Returns {perPost: [{interactions, denominator, basis, rate}], medianRate, aggregateRate, totalInteractions, totalDenominator}; rates are percentages. Returns {error: "invalid_input"} when a post is malformed or bases differ. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs93/100
Input documentation gives an agent enough context to populate the arguments.
posts › items › likes
Like or reaction count for the post.
Clear enough to supply this argument.
posts › items › comments
Comment or reply count for the post.
Clear enough to supply this argument.
posts › items › shares
Share, repost or retweet count for the post.
Clear enough to supply this argument.
posts › items › saves
Save or bookmark count for the post.
Clear enough to supply this argument.
posts › items › denominator
The audience number the rate is measured against: follower count, impressions or reach, as selected by basis.
Clear enough to supply this argument.
posts › items › basis
What the denominator counts. Use the same value for every post so rates are comparable.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "calculate_engagement",
"description": "CAPABILITY: `read`. Calculate engagement rates for supplied post metrics. This read-only, deterministic operation performs no mutation or network request. Returns {perPost: [{interactions, denominator, basis, rate}], medianRate, aggregateRate, totalInteractions, totalDenominator}; rates are percentages. Returns {error: \"invalid_input\"} when a post is malformed or bases differ. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"posts"
],
"properties": {
"posts": {
"type": "array",
"items": {
"type": "object",
"required": [
"denominator",
"basis"
],
"properties": {
"basis": {
"enum": [
"followers",
"impressions",
"reach"
],
"type": "string",
"description": "What the denominator counts. Use the same value for every post so rates are comparable."
},
"likes": {
"type": "number",
"description": "Like or reaction count for the post."
},
"saves": {
"type": "number",
"description": "Save or bookmark count for the post."
},
"shares": {
"type": "number",
"description": "Share, repost or retweet count for the post."
},
"comments": {
"type": "number",
"description": "Comment or reply count for the post."
},
"denominator": {
"type": "number",
"description": "The audience number the rate is measured against: follower count, impressions or reach, as selected by basis.",
"exclusiveMinimum": 0
}
}
},
"maxItems": 100,
"description": "Post metrics to score. Every post in one call must use the same basis."
}
}
},
"annotations": {
"title": "Calculate Engagement",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Return the authenticated TimeToPost user, bound organization and any org warning, plus THIS credential's capabilities (read / draft / publish) and canPublish. CALL THIS FIRST: a DRAFT-ONLY token can create drafts but is 403ed by every route that reaches a live audience (publish, schedule, approve, DM-send), so check canPublish before you plan any publishing step. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No input fields.
Annotations100/100
All four behavior hints are declared.
{
"title": "Show Current Account and Workspace",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
Output schema0/100
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "whoami",
"description": "Return the authenticated TimeToPost user, bound organization and any org warning, plus THIS credential's capabilities (read / draft / publish) and canPublish. CALL THIS FIRST: a DRAFT-ONLY token can create drafts but is 403ed by every route that reaches a live audience (publish, schedule, approve, DM-send), so check canPublish before you plan any publishing step. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "Show Current Account and Workspace",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
START HERE when mapping what TimeToPost can do. Returns the authoritative machine-readable capability map: post statuses + lifecycle, thread rules, per-platform platformData shapes, the engine-draft approval lifecycle and idempotency contract, the org/multi-account model (incl. the brand-vs-BrandVoice terminology distinction), metrics fields + ingestion cadence, AutoSEO, DM Funnels, the Weekly X Trend-Rider, the warm-outbound DM engager loop, backlink/email outreach, the link shortener + click analytics, Launch with TTP, the link-in-bio public profile, and THIS org's live connections including token-refresh semantics (e.g. X access tokens live 2h by design and refresh automatically at publish, a past expiresAt with hasRefreshCapability=true is normal, NOT a dead integration). Trust this over inferences from other tools' raw fields. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_capabilities",
"description": "START HERE when mapping what TimeToPost can do. Returns the authoritative machine-readable capability map: post statuses + lifecycle, thread rules, per-platform platformData shapes, the engine-draft approval lifecycle and idempotency contract, the org/multi-account model (incl. the brand-vs-BrandVoice terminology distinction), metrics fields + ingestion cadence, AutoSEO, DM Funnels, the Weekly X Trend-Rider, the warm-outbound DM engager loop, backlink/email outreach, the link shortener + click analytics, Launch with TTP, the link-in-bio public profile, and THIS org's live connections including token-refresh semantics (e.g. X access tokens live 2h by design and refresh automatically at publish, a past expiresAt with hasRefreshCapability=true is normal, NOT a dead integration). Trust this over inferences from other tools' raw fields. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "Get Capabilities",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description91/100
Return ranked, evidence-labeled prompt scaffolds for the active organization. Call this before generating any post or short and start from the first result. Use POST_HOOK for post openings and VIDEO for short-video structures. Organization winners rank above source-cited best-practice templates only when confidence is medium or high. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs83/100
Explain the meaning of this value, not just its name.
kind
Which kind of scaffold to return: POST_HOOK for the opening line of a post, VIDEO for the structure of a short video.
Clear enough to supply this argument.
niche
Optional niche used to prefer an exact or generic proven template.
Explain the meaning of this value, not just its name.
platform
Optional: only return winners and proven templates written for this platform.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "prompt_suggest",
"description": "Return ranked, evidence-labeled prompt scaffolds for the active organization. Call this before generating any post or short and start from the first result. Use POST_HOOK for post openings and VIDEO for short-video structures. Organization winners rank above source-cited best-practice templates only when confidence is medium or high. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"kind"
],
"properties": {
"kind": {
"enum": [
"POST_HOOK",
"VIDEO"
],
"type": "string",
"description": "Which kind of scaffold to return: POST_HOOK for the opening line of a post, VIDEO for the structure of a short video."
},
"niche": {
"type": "string",
"description": "Optional niche used to prefer an exact or generic proven template."
},
"platform": {
"enum": [
"instagram",
"twitter",
"youtube",
"linkedin",
"tiktok",
"facebook",
"wordpress",
"medium",
"pinterest",
"reddit",
"threads"
],
"type": "string",
"description": "Optional: only return winners and proven templates written for this platform."
}
}
},
"annotations": {
"title": "Suggest Prompt",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description90/100
List the active organization prompt library, including visible system and organization templates plus learned winner evidence. Use prompt_suggest for the ranked generation default and this tool to inspect the complete inventory. CAPABILITY: `read`: side-effect-free.
Describe the returned information so the agent knows whether it will answer the request.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "prompt_library_list",
"description": "List the active organization prompt library, including visible system and organization templates plus learned winner evidence. Use prompt_suggest for the ranked generation default and this tool to inspect the complete inventory. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "List Prompt Library",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
List the 6 TimeToPost free-tool offers with stable web URLs, honest free and gated tier summaries, and (where the free tier calls a backend endpoint) a public API path. This catalog call does not run a paid data provider. Use it to discover which teaser is free and what deeper report exists before calling a compute tool. MCP calls are authenticated and still pass through the public route rate limit. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_free_tools",
"description": "List the 6 TimeToPost free-tool offers with stable web URLs, honest free and gated tier summaries, and (where the free tier calls a backend endpoint) a public API path. This catalog call does not run a paid data provider. Use it to discover which teaser is free and what deeper report exists before calling a compute tool. MCP calls are authenticated and still pass through the public route rate limit. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "List Free Tools",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Audit a public URL for technical SEO and GEO readiness. The free tier uses a real server fetch, deterministic page checks, a 0 to 100 score, and plain-English fixes. The gated tier adds genuine live rankings and keyword gaps when a data source is connected, and reports an explicit unavailable or partial status otherwise. Authenticated MCP callers default to full=true; pass full=false for the free result. The request remains rate limited and never fabricates measured search data. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
url
Public HTTP or HTTPS page URL to audit. A bare hostname is accepted.
Clear enough to supply this argument.
full
Request the gated live search data report. Defaults to true for authenticated MCP callers; false requests the free heuristic tier.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "seo_audit",
"description": "Audit a public URL for technical SEO and GEO readiness. The free tier uses a real server fetch, deterministic page checks, a 0 to 100 score, and plain-English fixes. The gated tier adds genuine live rankings and keyword gaps when a data source is connected, and reports an explicit unavailable or partial status otherwise. Authenticated MCP callers default to full=true; pass full=false for the free result. The request remains rate limited and never fabricates measured search data. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"description": "Public HTTP or HTTPS page URL to audit. A bare hostname is accepted."
},
"full": {
"type": "boolean",
"description": "Request the gated live search data report. Defaults to true for authenticated MCP callers; false requests the free heuristic tier."
}
}
},
"annotations": {
"title": "SEO Audit",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description84/100
Return keyword opportunities for a seed keyword or site label. The free tier is explicitly directional: it uses an estimated provider, exposes coarse volume and difficulty bands, and redacts exact metrics. The gated tier requests genuine live volume, difficulty, CPC, intent, and trend data from a connected data source, and fails honestly when live data is not configured or unavailable. Authenticated MCP callers default to live=true; pass live=false for the free estimated tier. The request remains rate limited. CAPABILITY: `read`: side-effect-free.
Describe the returned information so the agent knows whether it will answer the request.
Inputs91/100
Input documentation gives an agent enough context to populate the arguments.
seed
Keyword, domain, or URL label to research.
Clear enough to supply this argument.
seedKind
Whether seed is a search phrase (keyword) or a domain or brand label (site). Site mode never crawls the site; it only estimates from the label.
Clear enough to supply this argument.
country
Optional market to pull search data for, as a two-letter ISO country code such as us or es. Defaults to us.
Clear enough to supply this argument.
live
Request the gated measured live data tier. Defaults to true for authenticated MCP callers; false requests estimated bands with exact metrics redacted.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "keyword_snapshot",
"description": "Return keyword opportunities for a seed keyword or site label. The free tier is explicitly directional: it uses an estimated provider, exposes coarse volume and difficulty bands, and redacts exact metrics. The gated tier requests genuine live volume, difficulty, CPC, intent, and trend data from a connected data source, and fails honestly when live data is not configured or unavailable. Authenticated MCP callers default to live=true; pass live=false for the free estimated tier. The request remains rate limited. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"seed",
"seedKind"
],
"properties": {
"live": {
"type": "boolean",
"description": "Request the gated measured live data tier. Defaults to true for authenticated MCP callers; false requests estimated bands with exact metrics redacted."
},
"seed": {
"type": "string",
"description": "Keyword, domain, or URL label to research."
},
"country": {
"type": "string",
"description": "Optional market to pull search data for, as a two-letter ISO country code such as us or es. Defaults to us."
},
"seedKind": {
"enum": [
"keyword",
"site"
],
"type": "string",
"description": "Whether seed is a search phrase (keyword) or a domain or brand label (site). Site mode never crawls the site; it only estimates from the label."
}
}
},
"annotations": {
"title": "Keyword Snapshot",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
Expand internal abbreviations and remove technical prefixes from the display title.
Description86/100
Check whether an AI model names a brand for a category. The free tier runs one source-labeled model probe. The gated tier runs multiple genuine probes, then derives visibility percentage, repeated competitors, and practical fixes from those responses. This is a point-in-time model check without live web search, not a universal visibility measurement. Authenticated MCP callers default to full=true; pass full=false for one free probe. The request remains rate limited and never invents citations. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs88/100
Input documentation gives an agent enough context to populate the arguments.
brand
Brand or product name to look for in the model response.
Clear enough to supply this argument.
category
Product or service category that a buyer might ask the model about.
Clear enough to supply this argument.
full
Request the gated multi-prompt report. Defaults to true for authenticated MCP callers; false requests one free probe.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "geo_visibility_check",
"description": "Check whether an AI model names a brand for a category. The free tier runs one source-labeled model probe. The gated tier runs multiple genuine probes, then derives visibility percentage, repeated competitors, and practical fixes from those responses. This is a point-in-time model check without live web search, not a universal visibility measurement. Authenticated MCP callers default to full=true; pass full=false for one free probe. The request remains rate limited and never invents citations. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"brand",
"category"
],
"properties": {
"full": {
"type": "boolean",
"description": "Request the gated multi-prompt report. Defaults to true for authenticated MCP callers; false requests one free probe."
},
"brand": {
"type": "string",
"description": "Brand or product name to look for in the model response."
},
"category": {
"type": "string",
"description": "Product or service category that a buyer might ask the model about."
}
}
},
"annotations": {
"title": "AI Search Visibility Check (GEO)",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Return the free, source-labeled public best-time curve for a platform. Results are aggregate relative scores by UTC hour and identify whether the source is first-party platform data or an editorial fallback; they are not the authenticated organization-specific model returned by get_optimal_times. timezone and days are forwarded as request hints, but callers must trust the response fields and must not claim a timezone conversion or calendar was generated when the endpoint returns only its UTC curve. This rate-limited free teaser does not create the gated downloadable seven-day platform and timezone calendar. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platform
Platform to return the public posting-time curve for.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
timezone
Optional IANA timezone request hint, for example America/New_York. The current response remains authoritative about whether conversion occurred.
Clear enough to supply this argument.
days
Optional number of days the caller wants to plan for. The teaser may still return only the hourly curve.
Clear enough to supply this argument.
Annotations100/100
All four behavior hints are declared.
{
"title": "Public Best Posting Times Preview",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
Output schema0/100
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "best_time_teaser",
"description": "Return the free, source-labeled public best-time curve for a platform. Results are aggregate relative scores by UTC hour and identify whether the source is first-party platform data or an editorial fallback; they are not the authenticated organization-specific model returned by get_optimal_times. timezone and days are forwarded as request hints, but callers must trust the response fields and must not claim a timezone conversion or calendar was generated when the endpoint returns only its UTC curve. This rate-limited free teaser does not create the gated downloadable seven-day platform and timezone calendar. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"platform"
],
"properties": {
"days": {
"type": "number",
"maximum": 7,
"minimum": 1,
"description": "Optional number of days the caller wants to plan for. The teaser may still return only the hourly curve."
},
"platform": {
"enum": [
"instagram",
"tiktok",
"twitter",
"linkedin",
"pinterest",
"youtube",
"facebook"
],
"type": "string",
"description": "Platform to return the public posting-time curve for."
},
"timezone": {
"type": "string",
"description": "Optional IANA timezone request hint, for example America/New_York. The current response remains authoritative about whether conversion occurred."
}
}
},
"annotations": {
"title": "Public Best Posting Times Preview",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
List connected social accounts (Instagram, TikTok, X, Pinterest, ...) and their status. Each account has an `id` (its integration id) and, if grouped, a `collectionId`/collection name. When a platform has multiple connected accounts, schedule_post REQUIRES those account `id`s in accountIds, a collection name/id, or explicit allAccounts:true. An implicit multi-account target is rejected. Kick requires exactly one explicitly selected account and its trusted metadata.broadcasterUserId for its own channel. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_integrations",
"description": "List connected social accounts (Instagram, TikTok, X, Pinterest, ...) and their status. Each account has an `id` (its integration id) and, if grouped, a `collectionId`/collection name. When a platform has multiple connected accounts, schedule_post REQUIRES those account `id`s in accountIds, a collection name/id, or explicit allAccounts:true. An implicit multi-account target is rejected. Kick requires exactly one explicitly selected account and its trusted metadata.broadcasterUserId for its own channel. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "List Integrations",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Read a connected TikTok account's CURRENT posting settings, straight from TikTok. Call this before every TikTok schedule_post: it returns who the post would go out as (creator_nickname/creator_username), privacy_level_options (the ONLY audiences this account may be offered, so never present one that is missing from this list), comment_disabled/duet_disabled/stitch_disabled (a disabled interaction MUST be sent as false in platformData.tiktok, and must not be offered to the human as a choice), and max_video_post_duration_sec (the longest video this account may upload). Show these to the human, get their answers, and send those answers. TikTok requires this lookup per posting session and forbids caching it between sessions, so do not reuse an earlier response. accountId is the integration id from list_integrations; omit it when only one TikTok account is connected. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
accountId
Integration id of the TikTok account (from list_integrations). Optional when only one TikTok account is connected.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_tiktok_creator_info",
"description": "Read a connected TikTok account's CURRENT posting settings, straight from TikTok. Call this before every TikTok schedule_post: it returns who the post would go out as (creator_nickname/creator_username), privacy_level_options (the ONLY audiences this account may be offered, so never present one that is missing from this list), comment_disabled/duet_disabled/stitch_disabled (a disabled interaction MUST be sent as false in platformData.tiktok, and must not be offered to the human as a choice), and max_video_post_duration_sec (the longest video this account may upload). Show these to the human, get their answers, and send those answers. TikTok requires this lookup per posting session and forbids caching it between sessions, so do not reuse an earlier response. accountId is the integration id from list_integrations; omit it when only one TikTok account is connected. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"accountId": {
"type": "string",
"description": "Integration id of the TikTok account (from list_integrations). Optional when only one TikTok account is connected."
}
}
},
"annotations": {
"title": "Get TikTok Creator Info",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Group a connected account into a collection (a named set of accounts, e.g. "acme" spanning its X + Instagram). CREATES the collection if the name is new, so this is also how you create one. Then schedule_post can post to the whole group via its `collection` field. Pass accountId (an integration id from list_integrations) and collection (the group name). Call list_integrations to see current groupings (each account shows its collection). To ungroup an account, pass an empty collection name. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
accountId
Integration id of the account to group (from list_integrations).
Clear enough to supply this argument.
collection
Collection name. Created if it does not exist yet. Empty string ungroups the account.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "add_to_collection",
"description": "Group a connected account into a collection (a named set of accounts, e.g. \"acme\" spanning its X + Instagram). CREATES the collection if the name is new, so this is also how you create one. Then schedule_post can post to the whole group via its `collection` field. Pass accountId (an integration id from list_integrations) and collection (the group name). Call list_integrations to see current groupings (each account shows its collection). To ungroup an account, pass an empty collection name. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"accountId",
"collection"
],
"properties": {
"accountId": {
"type": "string",
"description": "Integration id of the account to group (from list_integrations)."
},
"collection": {
"type": "string",
"description": "Collection name. Created if it does not exist yet. Empty string ungroups the account."
}
}
},
"annotations": {
"title": "Add To Collection",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Hand the user a link to connect (authorize) a social account. Connecting a platform is a browser OAuth popup flow gated by that platform's own login and consent screen, so it can only be completed by a signed-in human clicking through it in their own browser, never by an API call or an AI agent acting alone. Use this tool whenever the user asks to connect, add, authorize or reconnect a social account: call it, then relay the returned url and instructions to the user verbatim so they can open it themselves. Do not try to call any auth/OAuth endpoint directly on the user's behalf, and do not claim the account is connected until the user confirms it or list_integrations shows it as connected. The link deep-links into Settings, Integrations tab, and auto-opens the connect flow for that platform. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
provider
Platform the user wants to connect; the returned link opens that platform's connect flow in Settings.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "connect_account_link",
"description": "Hand the user a link to connect (authorize) a social account. Connecting a platform is a browser OAuth popup flow gated by that platform's own login and consent screen, so it can only be completed by a signed-in human clicking through it in their own browser, never by an API call or an AI agent acting alone. Use this tool whenever the user asks to connect, add, authorize or reconnect a social account: call it, then relay the returned url and instructions to the user verbatim so they can open it themselves. Do not try to call any auth/OAuth endpoint directly on the user's behalf, and do not claim the account is connected until the user confirms it or list_integrations shows it as connected. The link deep-links into Settings, Integrations tab, and auto-opens the connect flow for that platform. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"provider"
],
"properties": {
"provider": {
"enum": [
"writeas",
"listmonk",
"farcaster",
"dribbble",
"mastodon",
"pixelfed",
"discord",
"lemmy",
"twitch",
"slack",
"kick",
"instagram",
"twitter",
"tiktok",
"facebook",
"pinterest",
"youtube",
"linkedin",
"linkedin_page",
"reddit",
"threads",
"tumblr"
],
"type": "string",
"description": "Platform the user wants to connect; the returned link opens that platform's connect flow in Settings."
}
}
},
"annotations": {
"title": "Connect Account Link",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description92/100
List eligible public communities hosted on one connected Lemmy API v3 instance. Returns communities (use the id as communityId) and nextCursor when more pages exist; follow it. Schedule with accountIds containing accountId and platformData.lemmy containing integrationId: accountId, communityId, title, and optional url. Connect manually in settings; never send session tokens through chat or MCP. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
accountId
Lemmy integration id from list_integrations.
Clear enough to supply this argument.
cursor
Pagination cursor: the nextCursor from the previous page. Omit for the first page.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_lemmy_communities",
"description": "List eligible public communities hosted on one connected Lemmy API v3 instance. Returns communities (use the id as communityId) and nextCursor when more pages exist; follow it. Schedule with accountIds containing accountId and platformData.lemmy containing integrationId: accountId, communityId, title, and optional url. Connect manually in settings; never send session tokens through chat or MCP. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"accountId"
],
"properties": {
"cursor": {
"type": "string",
"description": "Pagination cursor: the nextCursor from the previous page. Omit for the first page."
},
"accountId": {
"type": "string",
"description": "Lemmy integration id from list_integrations."
}
}
},
"annotations": {
"title": "List Lemmy Communities",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
List joined channels for one Slack integration in the active workspace. Returns channel ids and names plus nextCursor when more pages exist. Use a returned id as platformData.slack.channelId and bind platformData.slack.integrationId to accountId; include that accountId in accountIds when scheduling text or links. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
accountId
Slack integration id from list_integrations.
Clear enough to supply this argument.
cursor
Pagination cursor: the nextCursor from the previous page. Omit for the first page.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_slack_channels",
"description": "List joined channels for one Slack integration in the active workspace. Returns channel ids and names plus nextCursor when more pages exist. Use a returned id as platformData.slack.channelId and bind platformData.slack.integrationId to accountId; include that accountId in accountIds when scheduling text or links. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"accountId"
],
"properties": {
"cursor": {
"type": "string",
"description": "Pagination cursor: the nextCursor from the previous page. Omit for the first page."
},
"accountId": {
"type": "string",
"description": "Slack integration id from list_integrations."
}
}
},
"annotations": {
"title": "List Slack Channels",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
List the boards (id, name, privacy, pin count) of a connected Pinterest account. Call this before schedule_post with a pinterest target: every pin needs platformData.pinterest.boardId, taken from the ids returned here. With one Pinterest account connected accountId can be omitted; with several, pass accountId. 404s if no Pinterest integration is connected. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
accountId
Pinterest integration id from list_integrations. Needed only when several Pinterest accounts are connected.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_boards",
"description": "List the boards (id, name, privacy, pin count) of a connected Pinterest account. Call this before schedule_post with a pinterest target: every pin needs platformData.pinterest.boardId, taken from the ids returned here. With one Pinterest account connected accountId can be omitted; with several, pass accountId. 404s if no Pinterest integration is connected. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"accountId": {
"type": "string",
"description": "Pinterest integration id from list_integrations. Needed only when several Pinterest accounts are connected."
}
}
},
"annotations": {
"title": "List Boards",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
List posts for the active org. Returns each post with id, content, platforms, status, scheduledAt and publishedAt; use an id with get_post for full detail. Optional status filter. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs75/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
status
Only return posts in this lifecycle status: draft, scheduled, published, failed or partial. Omit for all posts.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_posts",
"description": "List posts for the active org. Returns each post with id, content, platforms, status, scheduledAt and publishedAt; use an id with get_post for full detail. Optional status filter. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"status": {
"type": "string",
"description": "Only return posts in this lifecycle status: draft, scheduled, published, failed or partial. Omit for all posts."
}
}
},
"annotations": {
"title": "List Posts",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
Fetch a single post by id: content, platforms, status (DRAFT|SCHEDULED|PUBLISHING|PUBLISHED|FAILED|PARTIAL), scheduledAt/publishedAt, mediaUrls, and platformData including per-platform published post ids/urls, thread segments, and firstCommentResults (per account: posted with its url, failed with the error, or skipped). Use after publishing to confirm the post actually went out and to get platform permalinks. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs98/100
Input documentation gives an agent enough context to populate the arguments.
postId
Post id, from list_posts or the schedule_post response.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_post",
"description": "Fetch a single post by id: content, platforms, status (DRAFT|SCHEDULED|PUBLISHING|PUBLISHED|FAILED|PARTIAL), scheduledAt/publishedAt, mediaUrls, and platformData including per-platform published post ids/urls, thread segments, and firstCommentResults (per account: posted with its url, failed with the error, or skipped). Use after publishing to confirm the post actually went out and to get platform permalinks. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"postId"
],
"properties": {
"postId": {
"type": "string",
"description": "Post id, from list_posts or the schedule_post response."
}
}
},
"annotations": {
"title": "Get Post",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Upload a SMALL image/video inline and get a hosted public URL for schedule_post mediaUrls. Only practical for tiny files: it carries the whole file as base64 in the request, which is ~1.35x the byte size. For ANY real photo or video (anything over ~100KB) use prepare_media_upload + finalize_media_upload instead, that flow streams the bytes straight to storage and does not force a huge payload through this call. Instagram and TikTok REQUIRE hosted media (they cannot take a text-only post). Note: once a post using the media publishes, the file is auto-deleted 30 days later. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs86/100
Input documentation gives an agent enough context to populate the arguments.
filename
Original filename incl. extension, e.g. "day01.mp4", the extension is preserved on the stored object.
Clear enough to supply this argument.
contentType
MIME type, e.g. image/png or video/mp4.
Clear enough to supply this argument.
dataBase64
The file bytes, base64-encoded (no data: URI prefix).
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "upload_media",
"description": "Upload a SMALL image/video inline and get a hosted public URL for schedule_post mediaUrls. Only practical for tiny files: it carries the whole file as base64 in the request, which is ~1.35x the byte size. For ANY real photo or video (anything over ~100KB) use prepare_media_upload + finalize_media_upload instead, that flow streams the bytes straight to storage and does not force a huge payload through this call. Instagram and TikTok REQUIRE hosted media (they cannot take a text-only post). Note: once a post using the media publishes, the file is auto-deleted 30 days later. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"filename",
"contentType",
"dataBase64"
],
"properties": {
"filename": {
"type": "string",
"description": "Original filename incl. extension, e.g. \"day01.mp4\", the extension is preserved on the stored object."
},
"dataBase64": {
"type": "string",
"description": "The file bytes, base64-encoded (no data: URI prefix)."
},
"contentType": {
"type": "string",
"description": "MIME type, e.g. image/png or video/mp4."
}
}
},
"annotations": {
"title": "Upload Media",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
STEP 1 of the media upload flow (use this for any real image/video, see upload_media only for tiny files). Returns a short-lived PRESIGNED upload URL you PUT the raw file bytes to yourself, outside MCP, with NO credentials (the signature is in the URL). Then call finalize_media_upload to register it and get the hosted publicUrl for schedule_post mediaUrls.
Flow:
1. prepare_media_upload { filename, contentType, size } -> { key, uploadUrl, publicUrl }
2. PUT the bytes to uploadUrl, e.g. `curl -X PUT --data-binary @day01.mp4 -H "Content-Type: video/mp4" "<uploadUrl>"`. Do NOT base64-encode; send raw bytes.
3. finalize_media_upload { key, contentType, size } -> registers the MediaAsset.
4. Use the publicUrl in schedule_post mediaUrls.
size must be the exact byte size of the file. The uploadUrl expires shortly, so PUT immediately after preparing (prepare/PUT/finalize one file at a time rather than preparing many up front). CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
filename
Original filename incl. extension, e.g. "day01.mp4".
Clear enough to supply this argument.
contentType
MIME type of the file, e.g. image/jpeg or video/mp4. Must match the Content-Type you use for the PUT.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "prepare_media_upload",
"description": "STEP 1 of the media upload flow (use this for any real image/video, see upload_media only for tiny files). Returns a short-lived PRESIGNED upload URL you PUT the raw file bytes to yourself, outside MCP, with NO credentials (the signature is in the URL). Then call finalize_media_upload to register it and get the hosted publicUrl for schedule_post mediaUrls.\n\nFlow:\n1. prepare_media_upload { filename, contentType, size } -> { key, uploadUrl, publicUrl }\n2. PUT the bytes to uploadUrl, e.g. `curl -X PUT --data-binary @day01.mp4 -H \"Content-Type: video/mp4\" \"<uploadUrl>\"`. Do NOT base64-encode; send raw bytes.\n3. finalize_media_upload { key, contentType, size } -> registers the MediaAsset.\n4. Use the publicUrl in schedule_post mediaUrls.\n\nsize must be the exact byte size of the file. The uploadUrl expires shortly, so PUT immediately after preparing (prepare/PUT/finalize one file at a time rather than preparing many up front). CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"filename",
"contentType",
"size"
],
"properties": {
"size": {
"type": "number",
"description": "Exact file size in bytes."
},
"filename": {
"type": "string",
"description": "Original filename incl. extension, e.g. \"day01.mp4\"."
},
"contentType": {
"type": "string",
"description": "MIME type of the file, e.g. image/jpeg or video/mp4. Must match the Content-Type you use for the PUT."
}
}
},
"annotations": {
"title": "Prepare Media Upload",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
STEP 3 of the media upload flow: call AFTER you have PUT the raw bytes to the presigned uploadUrl from prepare_media_upload. Registers the object as a MediaAsset in the org media library and returns it (its publicUrl is what you pass to schedule_post mediaUrls). Pass the same key/contentType/size from step 1. Idempotent: re-finalizing the same key returns the existing asset. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "finalize_media_upload",
"description": "STEP 3 of the media upload flow: call AFTER you have PUT the raw bytes to the presigned uploadUrl from prepare_media_upload. Registers the object as a MediaAsset in the org media library and returns it (its publicUrl is what you pass to schedule_post mediaUrls). Pass the same key/contentType/size from step 1. Idempotent: re-finalizing the same key returns the existing asset. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"key",
"contentType",
"size"
],
"properties": {
"key": {
"type": "string",
"description": "The object key returned by prepare_media_upload."
},
"size": {
"type": "number",
"description": "Exact file size in bytes."
},
"contentType": {
"type": "string",
"description": "Same MIME type used in prepare_media_upload."
}
}
},
"annotations": {
"title": "Finalize Media Upload",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description93/100
Create a post and schedule it. scheduledAt (future ISO 8601 with timezone) schedules it; omit it to create a DRAFT. ACCOUNT TARGETING: when a requested platform has several connected accounts you MUST pass accountIds (integration ids from list_integrations), collection, or allAccounts:true, otherwise the call is rejected with 400. With accountIds, platforms may be omitted. TIKTOK REQUIRES platformData.tiktok, and every field in it must be an answer YOU GOT FROM THE HUMAN for this post: who can see it, whether comments, duet and stitch are allowed, and whether it promotes their own brand or a third party. NEVER invent, assume, default or re-use answers from a previous post: TikTok requires the creator to choose with nothing pre-selected, and audits us on it. An incomplete block is rejected with the missing field names. If the human is unavailable, tell them to compose the TikTok post in the TimeToPost composer. Other platforms in a mixed request still schedule. TikTok also needs mediaUrls and goes to ONE account per post. PRIVACY_HOLD: show Mel the returned warning and the exact text/images. Only after his explicit yes in this same chat, retry the unchanged post with force:true, the exact holdId, and ownerConfirmed set to Mel's exact words. Never invent or relay approval; secrets and PII cannot be forced. Platform rules (Pinterest boardId from list_boards, YouTube Shorts, Kick, Twitch, Dribbble, X threads) are on the matching fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft` when scheduledAt is omitted (creates a DRAFT that never goes out on its own); `publish` when scheduledAt is set, because the scheduler will push it to a live audience. A draft-scoped token that passes scheduledAt gets a 403, drop scheduledAt and hand the draft to a human.
The description explains the capability, when to use it, and what to expect.
Inputs89/100
70 of 72 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values.
content
Post text, or the pin description for Pinterest, the first tweet of an X thread, the email body for Listmonk, or the chat message for Kick and Twitch.
Clear enough to supply this argument.
platforms
Provider keys to post to, e.g. instagram, twitter, pinterest, tiktok, kick. Optional when accountIds is given, in which case they are derived from the accounts. tiktok additionally requires a complete platformData.tiktok block of choices made by the human.
Clear enough to supply this argument.
accountIds
Target SPECIFIC connected accounts by their integration id (get ids from list_integrations). Required for a multi-account platform unless collection or allAccounts:true is provided.
Clear enough to supply this argument.
collection
Post to a collection by name or id: expands to every connected account in that collection (across platforms). Mutually exclusive with accountIds; if both are given, accountIds wins. 404s if the collection name/id is unknown.
Clear enough to supply this argument.
allAccounts
Explicitly allow publishing to every connected account for each requested platform. Required when intentionally targeting all accounts on a platform that has more than one.
Clear enough to supply this argument.
brand
Brand/niche attribution tag for per-play rollups; not the AI BrandVoice profile
Clear enough to supply this argument.
createdBy
Required agent or lane name for attribution in protected workspaces.
Explain the meaning of this value, not just its name.
force
Retry only after showing Mel the returned warning and receiving his explicit yes.
Clear enough to supply this argument.
holdId
Exact holdId returned by PRIVACY_HOLD.
Clear enough to supply this argument.
ownerConfirmed
Mel's exact approving words from the current chat. Relayed approvals do not count.
Clear enough to supply this argument.
scheduledAt
Future ISO 8601 datetime with Z or an offset, e.g. 2026-10-12T09:00:00Z
Explain what happens when this optional value is omitted.
mediaUrls
Public URLs of images or video to attach, e.g. from upload_media or finalize_media_upload. Pinterest needs one image URL, TikTok a video or images, Instagram Stories one item.
Clear enough to supply this argument.
thread
X/Twitter only: tweets 2..N in order, published as replies chained to content (content is tweet 1).
Clear enough to supply this argument.
firstComment › anyOf › 1 › text
Not provided
Missing description. Explain what this value means and how to supply it.
firstComment › anyOf › 1 › networks
Not provided
Missing description. Explain what this value means and how to supply it.
platformData › listmonk › subject
Email subject line of the campaign.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › listmonk › contentType
How content is interpreted in the email body.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › dribbble › title
Shot title shown on Dribbble.
Clear enough to supply this argument.
platformData › writeas › title
Post title shown above the body.
Clear enough to supply this argument.
platformData › writeas › font
Typeface the post is rendered in.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › writeas › lang
Language of the post as a two-letter code, e.g. en.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › writeas › slug
Custom URL slug for the post.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › writeas › originalUrl
URL where the text first appeared; shown as an Originally published at link.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › mastodon › visibility
Defaults to followers only (private).
Clear enough to supply this argument.
platformData › pixelfed › visibility
Defaults to public.
Clear enough to supply this argument.
platformData › lemmy › integrationId
Lemmy integration id from list_integrations.
Clear enough to supply this argument.
platformData › lemmy › communityId
Numeric community id from list_lemmy_communities.
Clear enough to supply this argument.
platformData › lemmy › title
Post title.
Clear enough to supply this argument.
platformData › lemmy › url
Optional link the post points to.
Clear enough to supply this argument.
platformData › twitch › integrationId
Twitch integration id from list_integrations.
Clear enough to supply this argument.
platformData › twitch › broadcasterId
Twitch user id of the channel owner, from that integration metadata in list_integrations.
Clear enough to supply this argument.
platformData › twitch › sharedChatAcknowledged
Confirms the human understands the message also reaches Shared Chat session channels.
Clear enough to supply this argument.
platformData › youtube › privacyStatus
Defaults to private.
Clear enough to supply this argument.
platformData › youtube › short
Publish as a YouTube Short. There is no separate Shorts API: the video must be vertical or square and at most 3 minutes, otherwise it is refused. The URL is stored as youtube.com/shorts/<id>.
Clear enough to supply this argument.
platformData › youtube › title
Video title. Defaults to the first line of content.
Clear enough to supply this argument.
platformData › instagram › isStory
Post as a Story instead of a feed post.
Clear enough to supply this argument.
platformData › instagram › isTrialReel
Post as a trial reel, shown to non-followers first.
Clear enough to supply this argument.
platformData › instagram › trialGraduation
How a trial reel graduates to followers. Defaults to MANUAL.
Clear enough to supply this argument.
platformData › instagram › variantOf
Post id of the original this post is a variant of, for the UGC variants comparison.
Clear enough to supply this argument.
platformData › instagram › variantChanges
What differs from the original post.
Clear enough to supply this argument.
platformData › slack › integrationId
Slack integration id from list_integrations.
Clear enough to supply this argument.
platformData › slack › channelId
Slack channel id from list_slack_channels.
Clear enough to supply this argument.
platformData › kick › integrationId
Kick integration id from list_integrations.
Clear enough to supply this argument.
platformData › kick › broadcasterUserId
metadata.broadcasterUserId of that integration in list_integrations. Never invent a channel id.
Clear enough to supply this argument.
platformData › pinterest › boardId
Required. Get valid ids from list_boards.
Clear enough to supply this argument.
platformData › pinterest › title
Pin title shown on the pin.
Clear enough to supply this argument.
platformData › pinterest › link
Destination URL the pin links to
Clear enough to supply this argument.
platformData › pinterest › altText
Accessibility alt text describing the image.
Clear enough to supply this argument.
platformData › pinterest › dominantColor
Optional hex color hint, e.g. "#6B4F3B"
Clear enough to supply this argument.
platformData › tiktok › privacy
Who can see the post, chosen by the human: PUBLIC_TO_EVERYONE (Everyone), MUTUAL_FOLLOW_FRIENDS (Friends), FOLLOWER_OF_CREATOR (Followers), SELF_ONLY (Only you). Must be one the account is actually offered.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
platformData › tiktok › allowComments
Did the human say comments are allowed? Send false if they did not say yes.
Clear enough to supply this argument.
platformData › tiktok › allowDuet
Did the human say Duet is allowed? Videos only; send false for photo posts.
Clear enough to supply this argument.
platformData › tiktok › allowStitch
Did the human say Stitch is allowed? Videos only; send false for photo posts.
Clear enough to supply this argument.
platformData › tiktok › yourBrand
Human declared the post promotes themselves or their own business (TikTok labels it Promotional content).
Clear enough to supply this argument.
platformData › tiktok › brandedContent
Human declared the post promotes a third party (TikTok labels it Paid partnership). Cannot be combined with privacy SELF_ONLY.
Clear enough to supply this argument.
dmFunnel › keywords
Comment keywords, e.g. ["CHECKLIST"]. Case, width and apostrophes are ignored.
Clear enough to supply this argument.
dmFunnel › dmTemplate
DM text written by the human. {{link}} marks where the link goes; without it the link is appended.
Clear enough to supply this argument.
dmFunnel › leadUrl
HTTPS link sent in the DM. With trackClicks false, this exact URL is sent.
Clear enough to supply this argument.
dmFunnel › publicReplyTemplate
Optional public reply under the comment, e.g. "Sent you a DM!".
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › integrationId
Instagram integration id; only needed when the post goes to several Instagram accounts.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › name
Human label for the funnel in dashboards. Defaults to a generated name.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › dailyCap
Most DMs this funnel sends per day, to protect the account.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › promoCode
Optional active Stripe promotion code, appended as ?promo=.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › utmCampaign
Value for the utm_campaign parameter appended to leadUrl.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › trackClicks
Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "schedule_post",
"description": "Create a post and schedule it. scheduledAt (future ISO 8601 with timezone) schedules it; omit it to create a DRAFT. ACCOUNT TARGETING: when a requested platform has several connected accounts you MUST pass accountIds (integration ids from list_integrations), collection, or allAccounts:true, otherwise the call is rejected with 400. With accountIds, platforms may be omitted. TIKTOK REQUIRES platformData.tiktok, and every field in it must be an answer YOU GOT FROM THE HUMAN for this post: who can see it, whether comments, duet and stitch are allowed, and whether it promotes their own brand or a third party. NEVER invent, assume, default or re-use answers from a previous post: TikTok requires the creator to choose with nothing pre-selected, and audits us on it. An incomplete block is rejected with the missing field names. If the human is unavailable, tell them to compose the TikTok post in the TimeToPost composer. Other platforms in a mixed request still schedule. TikTok also needs mediaUrls and goes to ONE account per post. PRIVACY_HOLD: show Mel the returned warning and the exact text/images. Only after his explicit yes in this same chat, retry the unchanged post with force:true, the exact holdId, and ownerConfirmed set to Mel's exact words. Never invent or relay approval; secrets and PII cannot be forced. Platform rules (Pinterest boardId from list_boards, YouTube Shorts, Kick, Twitch, Dribbble, X threads) are on the matching fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft` when scheduledAt is omitted (creates a DRAFT that never goes out on its own); `publish` when scheduledAt is set, because the scheduler will push it to a live audience. A draft-scoped token that passes scheduledAt gets a 403, drop scheduledAt and hand the draft to a human.",
"inputSchema": {
"type": "object",
"required": [
"content",
"platforms"
],
"properties": {
"brand": {
"type": "string",
"description": "Brand/niche attribution tag for per-play rollups; not the AI BrandVoice profile"
},
"force": {
"type": "boolean",
"description": "Retry only after showing Mel the returned warning and receiving his explicit yes."
},
"holdId": {
"type": "string",
"description": "Exact holdId returned by PRIVACY_HOLD."
},
"thread": {
"type": "array",
"items": {
"type": "string"
},
"description": "X/Twitter only: tweets 2..N in order, published as replies chained to content (content is tweet 1)."
},
"content": {
"type": "string",
"description": "Post text, or the pin description for Pinterest, the first tweet of an X thread, the email body for Listmonk, or the chat message for Kick and Twitch."
},
"dmFunnel": {
"type": "object",
"required": [
"keywords",
"dmTemplate",
"leadUrl"
],
"properties": {
"name": {
"type": "string",
"maxLength": 100,
"description": "Human label for the funnel in dashboards. Defaults to a generated name."
},
"active": {
"type": "boolean",
"description": "false = save paused. Default true."
},
"leadUrl": {
"type": "string",
"description": "HTTPS link sent in the DM. With trackClicks false, this exact URL is sent."
},
"dailyCap": {
"type": "number",
"maximum": 50,
"minimum": 1,
"description": "Most DMs this funnel sends per day, to protect the account."
},
"keywords": {
"type": "array",
"items": {
"type": "string",
"maxLength": 64,
"minLength": 1
},
"maxItems": 20,
"minItems": 1,
"description": "Comment keywords, e.g. [\"CHECKLIST\"]. Case, width and apostrophes are ignored."
},
"linkTags": {
"type": "object",
"properties": {
"note": {
"type": "string",
"maxLength": 200,
"description": "Free-text note stored with the short link."
},
"slug": {
"type": "string",
"pattern": "^[A-Za-z0-9][A-Za-z0-9_-]{0,63}$",
"description": "Readable short-link slug. Default <product>-<handle>-<post>."
},
"channel": {
"type": "string",
"maxLength": 32,
"description": "Platform label, default instagram."
},
"campaign": {
"type": "string",
"maxLength": 60,
"description": "Product or campaign label."
},
"postLabel": {
"type": "string",
"maxLength": 80,
"description": "Post id or label."
},
"accountHandle": {
"type": "string",
"maxLength": 80,
"description": "Account handle the post runs on."
}
},
"description": "Farm tags for the tracked go short link (https://go.<product>/<slug>).",
"additionalProperties": false
},
"promoCode": {
"type": "string",
"description": "Optional active Stripe promotion code, appended as ?promo=."
},
"dmTemplate": {
"type": "string",
"maxLength": 1000,
"minLength": 1,
"description": "DM text written by the human. {{link}} marks where the link goes; without it the link is appended."
},
"trackClicks": {
"type": "boolean",
"description": "Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters."
},
"utmCampaign": {
"type": "string",
"description": "Value for the utm_campaign parameter appended to leadUrl."
},
"integrationId": {
"type": "string",
"description": "Instagram integration id; only needed when the post goes to several Instagram accounts."
},
"publicReplyTemplate": {
"type": [
"string",
"null"
],
"maxLength": 500,
"description": "Optional public reply under the comment, e.g. \"Sent you a DM!\"."
}
},
"description": "Instagram comment-to-DM funnel created WITH this post (needs an Instagram target and comment management on the workspace). When someone comments a keyword, they get dmTemplate as a private reply (and publicReplyTemplate under their comment, if set). It is ON by default and goes live the moment the post publishes (within seconds of each comment); active:false saves it paused. Switching it on needs a publish-capable token; a draft-only token saves it paused. The response carries dmFunnel.status (live | waiting_for_publish | paused) and dmFunnelError when it could not be switched on.",
"additionalProperties": false
},
"createdBy": {
"type": "string",
"description": "Required agent or lane name for attribution in protected workspaces."
},
"mediaUrls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Public URLs of images or video to attach, e.g. from upload_media or finalize_media_upload. Pinterest needs one image URL, TikTok a video or images, Instagram Stories one item."
},
"platforms": {
"type": "array",
"items": {
"type": "string"
},
"description": "Provider keys to post to, e.g. instagram, twitter, pinterest, tiktok, kick. Optional when accountIds is given, in which case they are derived from the accounts. tiktok additionally requires a complete platformData.tiktok block of choices made by the human."
},
"accountIds": {
"type": "array",
"items": {
"type": "string"
},
"description": "Target SPECIFIC connected accounts by their integration id (get ids from list_integrations). Required for a multi-account platform unless collection or allAccounts:true is provided."
},
"collection": {
"type": "string",
"description": "Post to a collection by name or id: expands to every connected account in that collection (across platforms). Mutually exclusive with accountIds; if both are given, accountIds wins. 404s if the collection name/id is unknown."
},
"allAccounts": {
"type": "boolean",
"description": "Explicitly allow publishing to every connected account for each requested platform. Required when intentionally targeting all accounts on a platform that has more than one."
},
"scheduledAt": {
"type": "string",
"description": "Future ISO 8601 datetime with Z or an offset, e.g. 2026-10-12T09:00:00Z"
},
"firstComment": {
"anyOf": [
{
"type": "string",
"maxLength": 8000
},
{
"type": "object",
"properties": {
"text": {
"type": "string",
"maxLength": 8000
},
"networks": {
"type": "object",
"additionalProperties": {
"type": "string",
"maxLength": 8000
}
}
},
"additionalProperties": false
}
],
"description": "First comment posted under the post immediately after it publishes, written by the human (never add hashtags or text they did not write). Supported on X (as your reply), Bluesky, Mastodon, Pixelfed, Telegram and Slack; Instagram, Threads and Facebook only once the account has the comment permission; ignored on other networks. A string, or {text, networks:{twitter:'...', bluesky:'...'}} for per-network text (\"\" skips a network). A failed comment never fails the post; get_post shows platformData.firstCommentResults."
},
"platformData": {
"type": "object",
"properties": {
"kick": {
"type": "object",
"required": [
"integrationId",
"broadcasterUserId"
],
"properties": {
"integrationId": {
"type": "string",
"minLength": 1,
"description": "Kick integration id from list_integrations."
},
"broadcasterUserId": {
"type": "integer",
"maximum": 9007199254740991,
"minimum": 1,
"description": "metadata.broadcasterUserId of that integration in list_integrations. Never invent a channel id."
}
},
"description": "Kick chat: one plain-text message (at most 500 grapheme clusters and 2048 UTF-8 bytes) sent as the connected customer to their own channel only. No feed posts, media, threads or arbitrary audiences. Select exactly one Kick integration in accountIds. An uncertain send is held for review and never resubmitted automatically.",
"additionalProperties": false
},
"lemmy": {
"type": "object",
"required": [
"integrationId",
"communityId",
"title"
],
"properties": {
"url": {
"type": "string",
"description": "Optional link the post points to."
},
"title": {
"type": "string",
"maxLength": 200,
"minLength": 1,
"description": "Post title."
},
"communityId": {
"type": "integer",
"minimum": 1,
"description": "Numeric community id from list_lemmy_communities."
},
"integrationId": {
"type": "string",
"description": "Lemmy integration id from list_integrations."
}
},
"description": "Lemmy post target. Get the community id from list_lemmy_communities.",
"additionalProperties": false
},
"slack": {
"type": "object",
"required": [
"integrationId",
"channelId"
],
"properties": {
"channelId": {
"type": "string",
"description": "Slack channel id from list_slack_channels."
},
"integrationId": {
"type": "string",
"description": "Slack integration id from list_integrations."
}
},
"description": "Slack channel target. Get channel ids from list_slack_channels.",
"additionalProperties": false
},
"tiktok": {
"type": "object",
"required": [
"privacy",
"allowComments",
"allowDuet",
"allowStitch",
"yourBrand",
"brandedContent"
],
"properties": {
"privacy": {
"enum": [
"PUBLIC_TO_EVERYONE",
"MUTUAL_FOLLOW_FRIENDS",
"FOLLOWER_OF_CREATOR",
"SELF_ONLY"
],
"type": "string",
"description": "Who can see the post, chosen by the human: PUBLIC_TO_EVERYONE (Everyone), MUTUAL_FOLLOW_FRIENDS (Friends), FOLLOWER_OF_CREATOR (Followers), SELF_ONLY (Only you). Must be one the account is actually offered."
},
"allowDuet": {
"type": "boolean",
"description": "Did the human say Duet is allowed? Videos only; send false for photo posts."
},
"yourBrand": {
"type": "boolean",
"description": "Human declared the post promotes themselves or their own business (TikTok labels it Promotional content)."
},
"allowStitch": {
"type": "boolean",
"description": "Did the human say Stitch is allowed? Videos only; send false for photo posts."
},
"allowComments": {
"type": "boolean",
"description": "Did the human say comments are allowed? Send false if they did not say yes."
},
"brandedContent": {
"type": "boolean",
"description": "Human declared the post promotes a third party (TikTok labels it Paid partnership). Cannot be combined with privacy SELF_ONLY."
}
},
"description": "REQUIRED when tiktok is a target platform, and every value must come from the human, asked for this post. Do not invent them, do not carry them over from a previous post, and do not pick a \"sensible default\": TikTok requires the creator to choose these with nothing pre-selected. All six fields are mandatory; anything missing is rejected by name. Call get_tiktok_creator_info first to see which audiences this account may pick and whether the account has comments, duet or stitch switched off (a switched-off interaction must be sent as false)."
},
"twitch": {
"type": "object",
"required": [
"integrationId",
"broadcasterId",
"sharedChatAcknowledged"
],
"properties": {
"broadcasterId": {
"type": "string",
"description": "Twitch user id of the channel owner, from that integration metadata in list_integrations."
},
"integrationId": {
"type": "string",
"description": "Twitch integration id from list_integrations."
},
"sharedChatAcknowledged": {
"type": "boolean",
"const": true,
"description": "Confirms the human understands the message also reaches Shared Chat session channels."
}
},
"description": "Text chat to the selected account’s own channel, max 500 characters, no media or threads. User tokens also send to every Shared Chat session channel; acknowledge this explicitly. Not a feed post or video upload.",
"additionalProperties": false
},
"writeas": {
"type": "object",
"properties": {
"font": {
"enum": [
"serif",
"sans",
"wrap",
"mono",
"code"
],
"type": "string",
"description": "Typeface the post is rendered in."
},
"lang": {
"type": "string",
"pattern": "^[a-zA-Z]{2}$",
"description": "Language of the post as a two-letter code, e.g. en."
},
"slug": {
"type": "string",
"pattern": "^[a-z0-9][a-z0-9-]{0,99}$",
"description": "Custom URL slug for the post."
},
"title": {
"type": "string",
"maxLength": 255,
"description": "Post title shown above the body."
},
"originalUrl": {
"type": "string",
"maxLength": 2048,
"description": "URL where the text first appeared; shown as an Originally published at link."
}
},
"description": "Optional for Write.as and WriteFreely blogs. originalUrl adds a visible Originally published at link: the API has no canonical tag field.",
"additionalProperties": false
},
"youtube": {
"type": "object",
"properties": {
"short": {
"type": "boolean",
"description": "Publish as a YouTube Short. There is no separate Shorts API: the video must be vertical or square and at most 3 minutes, otherwise it is refused. The URL is stored as youtube.com/shorts/<id>."
},
"title": {
"type": "string",
"maxLength": 100,
"description": "Video title. Defaults to the first line of content."
},
"privacyStatus": {
"enum": [
"private",
"unlisted",
"public"
],
"type": "string",
"description": "Defaults to private."
}
},
"description": "YouTube video options. #Shorts is appended to the description when short is true and it is missing.",
"additionalProperties": false
},
"dribbble": {
"type": "object",
"required": [
"title"
],
"properties": {
"title": {
"type": "string",
"pattern": "\\S",
"maxLength": 200,
"minLength": 1,
"description": "Shot title shown on Dribbble."
}
},
"description": "Dribbble shot. Needs exactly one GIF, JPG or PNG image of 400x300 or 800x600 pixels (max 8MB) in mediaUrls and one account with image-upload permission. No text-only posts, videos or threads. Uploads may still be processing; uncertain delivery holds automatic retries.",
"additionalProperties": false
},
"listmonk": {
"type": "object",
"required": [
"subject"
],
"properties": {
"subject": {
"type": "string",
"pattern": "^[^\\r\\n]+$",
"maxLength": 998,
"minLength": 1,
"description": "Email subject line of the campaign."
},
"contentType": {
"enum": [
"plain",
"html"
],
"type": "string",
"default": "plain",
"description": "How content is interpreted in the email body."
}
},
"description": "Required for Listmonk: content is the email body; select one mailing list via accountIds. No attachments or threads.",
"additionalProperties": false
},
"mastodon": {
"type": "object",
"properties": {
"visibility": {
"enum": [
"private",
"unlisted",
"public"
],
"type": "string",
"description": "Defaults to followers only (private)."
}
},
"description": "Mastodon options.",
"additionalProperties": false
},
"pixelfed": {
"type": "object",
"properties": {
"visibility": {
"enum": [
"private",
"unlisted",
"public"
],
"type": "string",
"description": "Defaults to public."
}
},
"description": "Pixelfed options.",
"additionalProperties": false
},
"instagram": {
"type": "object",
"properties": {
"isStory": {
"type": "boolean",
"description": "Post as a Story instead of a feed post."
},
"variantOf": {
"type": "string",
"maxLength": 120,
"description": "Post id of the original this post is a variant of, for the UGC variants comparison."
},
"isTrialReel": {
"type": "boolean",
"description": "Post as a trial reel, shown to non-followers first."
},
"variantChanges": {
"type": "array",
"items": {
"enum": [
"text_position",
"hook",
"audio"
],
"type": "string"
},
"maxItems": 3,
"description": "What differs from the original post."
},
"trialGraduation": {
"enum": [
"MANUAL",
"SS_PERFORMANCE"
],
"type": "string",
"description": "How a trial reel graduates to followers. Defaults to MANUAL."
}
},
"description": "Instagram options. isStory=true posts as a Story: exactly one image or video, no caption, no first comment, no DM funnel. isTrialReel=true posts a trial reel (shown to non-followers first; exactly one video; trialGraduation MANUAL by default or SS_PERFORMANCE). The API cannot report trial status later, so it is recorded at post time. variantOf = id of the original post and variantChanges (text_position|hook|audio) feed the UGC variants comparison.",
"additionalProperties": false
},
"pinterest": {
"type": "object",
"required": [
"boardId"
],
"properties": {
"link": {
"type": "string",
"description": "Destination URL the pin links to"
},
"title": {
"type": "string",
"description": "Pin title shown on the pin."
},
"altText": {
"type": "string",
"description": "Accessibility alt text describing the image."
},
"boardId": {
"type": "string",
"description": "Required. Get valid ids from list_boards."
},
"dominantColor": {
"type": "string",
"description": "Optional hex color hint, e.g. \"#6B4F3B\""
}
},
"description": "Required when pinterest is a target platform. mediaUrls must hold exactly one image URL; content becomes the pin description."
}
},
"description": "Platform-specific fields: pinterest pin fields, and the TikTok posting choices."
},
"ownerConfirmed": {
"type": "string",
"description": "Mel's exact approving words from the current chat. Relayed approvals do not count."
}
}
},
"annotations": {
"title": "Schedule Post",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
Queue an existing draft/scheduled/failed post for immediate publishing on the next scheduler tick. Uses the real scheduler/provider pipeline; it does not mark records published until the platform publish succeeds. Returns the queued post; follow with get_post to confirm it went out. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
postId
Id of the post to publish now, from list_posts or schedule_post.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "publish_post",
"description": "Queue an existing draft/scheduled/failed post for immediate publishing on the next scheduler tick. Uses the real scheduler/provider pipeline; it does not mark records published until the platform publish succeeds. Returns the queued post; follow with get_post to confirm it went out. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"postId"
],
"properties": {
"postId": {
"type": "string",
"description": "Id of the post to publish now, from list_posts or schedule_post."
}
}
},
"annotations": {
"title": "Publish Post",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
Shorten the title to an action and subject; move usage details into the description.
Description90/100
Edit a draft or scheduled post IN PLACE: same post id, no delete and re-create. Send only what changes. scheduledAt moves it to a new future time (ISO 8601 with Z or an offset such as +02:00) and also schedules a draft; draft=true takes it out of the queue. content replaces the text, mediaUrls replaces the media list, accountIds replaces the target accounts (ids from list_integrations; an account may be added only on networks that need no extra per-post choices, such as X, LinkedIn, Facebook, Instagram, Threads, Bluesky, Mastodon, Telegram, Discord, Farcaster). The same per-network limits as schedule_post apply. Published, publishing and failed posts cannot be edited (409). The old time can no longer fire once the call returns; the post publishes at the new time. Moving a draft into the queue counts toward the plan post limit; moving an already scheduled post does not. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs80/100
23 of 25 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values.
postId
Post id (from schedule_post or list_posts)
Clear enough to supply this argument.
scheduledAt
New future time, ISO 8601 with Z or an offset such as +02:00
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
draft
true: unschedule it and keep it as a draft
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
content
Replacement text/caption
Clear enough to supply this argument.
mediaUrls
Replacement media list ([] removes all media)
Clear enough to supply this argument.
accountIds
Replacement target account ids from list_integrations
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
firstComment › anyOf › 2 › text
Not provided
Missing description. Explain what this value means and how to supply it.
firstComment › anyOf › 2 › networks
Not provided
Missing description. Explain what this value means and how to supply it.
dmFunnel › anyOf › 0 › keywords
Comment keywords, e.g. ["CHECKLIST"]. Case, width and apostrophes are ignored.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › anyOf › 0 › dmTemplate
DM text written by the human. {{link}} marks where the link goes; without it the link is appended.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › anyOf › 0 › leadUrl
HTTPS link sent in the DM. With trackClicks false, this exact URL is sent.
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › publicReplyTemplate
Optional public reply under the comment, e.g. "Sent you a DM!".
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › integrationId
Instagram integration id; only needed when the post goes to several Instagram accounts.
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › name
Human label for the funnel in dashboards. Defaults to a generated name.
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › dailyCap
Most DMs this funnel sends per day, to protect the account.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmFunnel › anyOf › 0 › promoCode
Optional active Stripe promotion code, appended as ?promo=.
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › utmCampaign
Value for the utm_campaign parameter appended to leadUrl.
Clear enough to supply this argument.
dmFunnel › anyOf › 0 › trackClicks
Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "update_post",
"description": "Edit a draft or scheduled post IN PLACE: same post id, no delete and re-create. Send only what changes. scheduledAt moves it to a new future time (ISO 8601 with Z or an offset such as +02:00) and also schedules a draft; draft=true takes it out of the queue. content replaces the text, mediaUrls replaces the media list, accountIds replaces the target accounts (ids from list_integrations; an account may be added only on networks that need no extra per-post choices, such as X, LinkedIn, Facebook, Instagram, Threads, Bluesky, Mastodon, Telegram, Discord, Farcaster). The same per-network limits as schedule_post apply. Published, publishing and failed posts cannot be edited (409). The old time can no longer fire once the call returns; the post publishes at the new time. Moving a draft into the queue counts toward the plan post limit; moving an already scheduled post does not. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"postId"
],
"properties": {
"draft": {
"type": "boolean",
"const": true,
"description": "true: unschedule it and keep it as a draft"
},
"postId": {
"type": "string",
"description": "Post id (from schedule_post or list_posts)"
},
"content": {
"type": "string",
"description": "Replacement text/caption"
},
"dmFunnel": {
"anyOf": [
{
"type": "object",
"required": [
"keywords",
"dmTemplate",
"leadUrl"
],
"properties": {
"name": {
"type": "string",
"maxLength": 100,
"description": "Human label for the funnel in dashboards. Defaults to a generated name."
},
"active": {
"type": "boolean",
"description": "false = save paused. Default true."
},
"leadUrl": {
"type": "string",
"description": "HTTPS link sent in the DM. With trackClicks false, this exact URL is sent."
},
"dailyCap": {
"type": "number",
"maximum": 50,
"minimum": 1,
"description": "Most DMs this funnel sends per day, to protect the account."
},
"keywords": {
"type": "array",
"items": {
"type": "string",
"maxLength": 64,
"minLength": 1
},
"maxItems": 20,
"minItems": 1,
"description": "Comment keywords, e.g. [\"CHECKLIST\"]. Case, width and apostrophes are ignored."
},
"linkTags": {
"type": "object",
"properties": {
"note": {
"type": "string",
"maxLength": 200,
"description": "Free-text note stored with the short link."
},
"slug": {
"type": "string",
"pattern": "^[A-Za-z0-9][A-Za-z0-9_-]{0,63}$",
"description": "Readable short-link slug. Default <product>-<handle>-<post>."
},
"channel": {
"type": "string",
"maxLength": 32,
"description": "Platform label, default instagram."
},
"campaign": {
"type": "string",
"maxLength": 60,
"description": "Product or campaign label."
},
"postLabel": {
"type": "string",
"maxLength": 80,
"description": "Post id or label."
},
"accountHandle": {
"type": "string",
"maxLength": 80,
"description": "Account handle the post runs on."
}
},
"description": "Farm tags for the tracked go short link (https://go.<product>/<slug>).",
"additionalProperties": false
},
"promoCode": {
"type": "string",
"description": "Optional active Stripe promotion code, appended as ?promo=."
},
"dmTemplate": {
"type": "string",
"maxLength": 1000,
"minLength": 1,
"description": "DM text written by the human. {{link}} marks where the link goes; without it the link is appended."
},
"trackClicks": {
"type": "boolean",
"description": "Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters."
},
"utmCampaign": {
"type": "string",
"description": "Value for the utm_campaign parameter appended to leadUrl."
},
"integrationId": {
"type": "string",
"description": "Instagram integration id; only needed when the post goes to several Instagram accounts."
},
"publicReplyTemplate": {
"type": [
"string",
"null"
],
"maxLength": 500,
"description": "Optional public reply under the comment, e.g. \"Sent you a DM!\"."
}
},
"description": "Instagram comment-to-DM funnel created WITH this post (needs an Instagram target and comment management on the workspace). When someone comments a keyword, they get dmTemplate as a private reply (and publicReplyTemplate under their comment, if set). It is ON by default and goes live the moment the post publishes (within seconds of each comment); active:false saves it paused. Switching it on needs a publish-capable token; a draft-only token saves it paused. The response carries dmFunnel.status (live | waiting_for_publish | paused) and dmFunnelError when it could not be switched on.",
"additionalProperties": false
},
{
"type": "null"
}
],
"description": "Add or edit the post's Instagram DM funnel (same shape as schedule_post dmFunnel); null pauses it. Works on published posts too (goes live at once)."
},
"mediaUrls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Replacement media list ([] removes all media)"
},
"accountIds": {
"type": "array",
"items": {
"type": "string"
},
"minItems": 1,
"description": "Replacement target account ids from list_integrations"
},
"scheduledAt": {
"type": "string",
"description": "New future time, ISO 8601 with Z or an offset such as +02:00"
},
"firstComment": {
"anyOf": [
{
"type": "string",
"maxLength": 8000
},
{
"type": "null"
},
{
"type": "object",
"properties": {
"text": {
"type": "string",
"maxLength": 8000
},
"networks": {
"type": "object",
"additionalProperties": {
"type": "string",
"maxLength": 8000
}
}
},
"additionalProperties": false
}
],
"description": "Replacement first comment (same shape as schedule_post firstComment). null or \"\" removes it."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Update / Reschedule Post",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
Cancel a post by permanently deleting it (e.g. a scheduled one before it fires). NOT reversible: the post record is removed and cannot be restored, so to keep the text use update_post with draft=true instead. Does not remove anything already published on a platform. Returns {success, message}. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "cancel_post",
"description": "Cancel a post by permanently deleting it (e.g. a scheduled one before it fires). NOT reversible: the post record is removed and cannot be restored, so to keep the text use update_post with draft=true instead. Does not remove anything already published on a platform. Returns {success, message}. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"postId"
],
"properties": {
"postId": {
"type": "string",
"description": "Id of the post to delete, from list_posts."
}
}
},
"annotations": {
"title": "Cancel Post",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Aggregated engagement metrics (likes, comments, shares, views) for the active org. Returns org-wide totals; use get_post_metrics for individual posts. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Returns the org's best posting time slots (top hours and days, with the underlying sample size / confidence) computed from its OWN historical engagement, not industry averages. CALL THIS BEFORE SCHEDULING: when planning a batch of posts (e.g. a week of content), call get_optimal_times first, read the top per-platform slots, then sequence schedule_post calls so each post lands in one of those windows. Pass an optional platform to get slots for a single platform. If the org has little history, fall back to the public no-auth best-time curve at GET /api/analytics/public/best-times. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Input documentation gives an agent enough context to populate the arguments.
platform
Optional platform key, e.g. instagram, tiktok or twitter. Omit to get slots for every platform with enough history.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_optimal_times",
"description": "Returns the org's best posting time slots (top hours and days, with the underlying sample size / confidence) computed from its OWN historical engagement, not industry averages. CALL THIS BEFORE SCHEDULING: when planning a batch of posts (e.g. a week of content), call get_optimal_times first, read the top per-platform slots, then sequence schedule_post calls so each post lands in one of those windows. Pass an optional platform to get slots for a single platform. If the org has little history, fall back to the public no-auth best-time curve at GET /api/analytics/public/best-times. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"platform": {
"type": "string",
"description": "Optional platform key, e.g. instagram, tiktok or twitter. Omit to get slots for every platform with enough history."
}
}
},
"annotations": {
"title": "Get Optimal Times",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Health of the background post scheduler. Returns when it last ticked and recent failure counts; check it when scheduled posts are late. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
Submit a batch of platform drafts for one engine artifact (e.g. a digest issue). Same mechanism as create_drafts, kept under an older name. Drafts are ALWAYS created pending_approval and NEVER auto-publish: a human must approve_draft (or use the dashboard) before anything is scheduled. Idempotent on external_ref: resending a batch updates still-pending drafts, never resurrects approved or rejected drafts, and never overwrites a draft a human edited in the dashboard. Returns the created or updated drafts per variant. Per-field rules (platform aliases, thread segments, schedule_hint, account_ids) are on the variants fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs91/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
external_ref
Your own stable id for this batch (e.g. the digest id). Reusing it updates the same drafts instead of duplicating them.
Clear enough to supply this argument.
brand
Brand/niche attribution tag for per-play rollups (e.g. "emplaw-watchdog"); NOT the AI BrandVoice writing-style profile
Clear enough to supply this argument.
variants › items › platform
Provider key to draft for, e.g. twitter, linkedin, instagram, tiktok, wordpress, tumblr. The aliases x, x_post and x_thread all map to twitter.
Clear enough to supply this argument.
variants › items › segments
Post text. One element is a single post; two or more on twitter become a thread, one tweet per element, published as chained replies.
Clear enough to supply this argument.
variants › items › link
URL (e.g. an archive link) appended to the final segment at approval when not already present.
Clear enough to supply this argument.
variants › items › media › items › url
Public URL of the media file.
Clear enough to supply this argument.
variants › items › media › items › alt
Alt text describing the media for accessibility.
Clear enough to supply this argument.
variants › items › schedule_hint
When to publish after approval: "optimal" picks a time from the org's own best-times model at approval; otherwise a future ISO 8601 datetime with timezone. Omit to let the approver choose.
Clear enough to supply this argument.
variants › items › account_ids
Integration id(s) from list_integrations the approved post must target. REQUIRED when the org has more than one connected account on this platform, otherwise approval is refused as ambiguous.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "create_digest_drafts",
"description": "Submit a batch of platform drafts for one engine artifact (e.g. a digest issue). Same mechanism as create_drafts, kept under an older name. Drafts are ALWAYS created pending_approval and NEVER auto-publish: a human must approve_draft (or use the dashboard) before anything is scheduled. Idempotent on external_ref: resending a batch updates still-pending drafts, never resurrects approved or rejected drafts, and never overwrites a draft a human edited in the dashboard. Returns the created or updated drafts per variant. Per-field rules (platform aliases, thread segments, schedule_hint, account_ids) are on the variants fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"external_ref",
"variants"
],
"properties": {
"brand": {
"type": "string",
"description": "Brand/niche attribution tag for per-play rollups (e.g. \"emplaw-watchdog\"); NOT the AI BrandVoice writing-style profile"
},
"variants": {
"type": "array",
"items": {
"type": "object",
"required": [
"platform",
"segments"
],
"properties": {
"link": {
"type": "string",
"description": "URL (e.g. an archive link) appended to the final segment at approval when not already present."
},
"media": {
"type": "array",
"items": {
"type": "object",
"required": [
"url"
],
"properties": {
"alt": {
"type": "string",
"description": "Alt text describing the media for accessibility."
},
"url": {
"type": "string",
"description": "Public URL of the media file."
}
}
},
"description": "Images or video to attach to this draft."
},
"platform": {
"type": "string",
"description": "Provider key to draft for, e.g. twitter, linkedin, instagram, tiktok, wordpress, tumblr. The aliases x, x_post and x_thread all map to twitter."
},
"segments": {
"type": "array",
"items": {
"type": "string"
},
"description": "Post text. One element is a single post; two or more on twitter become a thread, one tweet per element, published as chained replies."
},
"account_ids": {
"type": "array",
"items": {
"type": "string"
},
"description": "Integration id(s) from list_integrations the approved post must target. REQUIRED when the org has more than one connected account on this platform, otherwise approval is refused as ambiguous."
},
"schedule_hint": {
"type": "string",
"description": "When to publish after approval: \"optimal\" picks a time from the org's own best-times model at approval; otherwise a future ISO 8601 datetime with timezone. Omit to let the approver choose."
},
"provider_metadata": {
"type": "object",
"properties": {
"pinterest": {
"type": "object",
"required": [
"boardId"
],
"properties": {
"link": {
"type": "string",
"description": "Destination URL the pin links to."
},
"title": {
"type": "string",
"description": "Pin title."
},
"altText": {
"type": "string",
"description": "Accessibility alt text describing the image."
},
"boardId": {
"type": "string",
"description": "Required; obtain from list_boards for the targeted account."
},
"dominantColor": {
"type": "string",
"description": "Optional hex color hint, e.g. \"#6B4F3B\"."
}
},
"description": "Pinterest pin fields."
}
},
"description": "Required for Pinterest. Provider fields preserved through human approval."
}
}
},
"description": "One entry per platform draft to create; each is reviewed and approved separately."
},
"external_ref": {
"type": "string",
"description": "Your own stable id for this batch (e.g. the digest id). Reusing it updates the same drafts instead of duplicating them."
}
}
},
"annotations": {
"title": "Create Digest Drafts",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
Call prompt_suggest first and base the prompt on the top result. Submit a batch of platform drafts for one engine artifact (a digest issue, a generated content batch, anything). Prefer this name for new integrations; create_digest_drafts is the same tool under an older name. Drafts are ALWAYS created pending_approval and NEVER auto-publish: a human must approve_draft (or use the dashboard) before anything is scheduled. Idempotent on external_ref: resending a batch updates still-pending drafts, never resurrects approved or rejected drafts, and never overwrites a draft a human edited in the dashboard. Returns the created or updated drafts per variant. Per-field rules (platform aliases, thread segments, schedule_hint, account_ids) are on the variants fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Explain the meaning of this value, not just its name.
external_ref
Your own stable id for this batch (e.g. the digest id). Reusing it updates the same drafts instead of duplicating them.
Clear enough to supply this argument.
brand
Brand/niche attribution tag for per-play rollups (e.g. "emplaw-watchdog"); NOT the AI BrandVoice writing-style profile
Clear enough to supply this argument.
variants › items › platform
Provider key to draft for, e.g. twitter, linkedin, instagram, tiktok, wordpress, tumblr. The aliases x, x_post and x_thread all map to twitter.
Clear enough to supply this argument.
variants › items › segments
Post text. One element is a single post; two or more on twitter become a thread, one tweet per element, published as chained replies.
Clear enough to supply this argument.
variants › items › link
URL (e.g. an archive link) appended to the final segment at approval when not already present.
Clear enough to supply this argument.
variants › items › media › items › url
Public URL of the media file.
Clear enough to supply this argument.
variants › items › media › items › alt
Alt text describing the media for accessibility.
Clear enough to supply this argument.
variants › items › schedule_hint
When to publish after approval: "optimal" picks a time from the org's own best-times model at approval; otherwise a future ISO 8601 datetime with timezone. Omit to let the approver choose.
Clear enough to supply this argument.
variants › items › account_ids
Integration id(s) from list_integrations the approved post must target. REQUIRED when the org has more than one connected account on this platform, otherwise approval is refused as ambiguous.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "create_drafts",
"description": "Call prompt_suggest first and base the prompt on the top result. Submit a batch of platform drafts for one engine artifact (a digest issue, a generated content batch, anything). Prefer this name for new integrations; create_digest_drafts is the same tool under an older name. Drafts are ALWAYS created pending_approval and NEVER auto-publish: a human must approve_draft (or use the dashboard) before anything is scheduled. Idempotent on external_ref: resending a batch updates still-pending drafts, never resurrects approved or rejected drafts, and never overwrites a draft a human edited in the dashboard. Returns the created or updated drafts per variant. Per-field rules (platform aliases, thread segments, schedule_hint, account_ids) are on the variants fields. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"external_ref",
"variants"
],
"properties": {
"brand": {
"type": "string",
"description": "Brand/niche attribution tag for per-play rollups (e.g. \"emplaw-watchdog\"); NOT the AI BrandVoice writing-style profile"
},
"variants": {
"type": "array",
"items": {
"type": "object",
"required": [
"platform",
"segments"
],
"properties": {
"link": {
"type": "string",
"description": "URL (e.g. an archive link) appended to the final segment at approval when not already present."
},
"media": {
"type": "array",
"items": {
"type": "object",
"required": [
"url"
],
"properties": {
"alt": {
"type": "string",
"description": "Alt text describing the media for accessibility."
},
"url": {
"type": "string",
"description": "Public URL of the media file."
}
}
},
"description": "Images or video to attach to this draft."
},
"platform": {
"type": "string",
"description": "Provider key to draft for, e.g. twitter, linkedin, instagram, tiktok, wordpress, tumblr. The aliases x, x_post and x_thread all map to twitter."
},
"segments": {
"type": "array",
"items": {
"type": "string"
},
"description": "Post text. One element is a single post; two or more on twitter become a thread, one tweet per element, published as chained replies."
},
"account_ids": {
"type": "array",
"items": {
"type": "string"
},
"description": "Integration id(s) from list_integrations the approved post must target. REQUIRED when the org has more than one connected account on this platform, otherwise approval is refused as ambiguous."
},
"schedule_hint": {
"type": "string",
"description": "When to publish after approval: \"optimal\" picks a time from the org's own best-times model at approval; otherwise a future ISO 8601 datetime with timezone. Omit to let the approver choose."
},
"provider_metadata": {
"type": "object",
"properties": {
"pinterest": {
"type": "object",
"required": [
"boardId"
],
"properties": {
"link": {
"type": "string",
"description": "Destination URL the pin links to."
},
"title": {
"type": "string",
"description": "Pin title."
},
"altText": {
"type": "string",
"description": "Accessibility alt text describing the image."
},
"boardId": {
"type": "string",
"description": "Required; obtain from list_boards for the targeted account."
},
"dominantColor": {
"type": "string",
"description": "Optional hex color hint, e.g. \"#6B4F3B\"."
}
},
"description": "Pinterest pin fields."
}
},
"description": "Required for Pinterest. Provider fields preserved through human approval."
}
}
},
"description": "One entry per platform draft to create; each is reviewed and approved separately."
},
"external_ref": {
"type": "string",
"description": "Your own stable id for this batch (e.g. the digest id). Reusing it updates the same drafts instead of duplicating them."
}
}
},
"annotations": {
"title": "Create Drafts",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Import a CSV of posts as DRAFTS ONLY. Nothing is scheduled or published. dry_run true validates each row and writes nothing: supported network, a connected account in this org, the per-network character limit, the date, and media URL syntax. dry_run false creates pending drafts through the same path as create_drafts. Columns: text, scheduled_at, timezone, network, account, media_urls, first_comment, link, tags. Header names are case-insensitive. Also accepts calendar export headers (Message, Social network, Social profile, Date, Time) and a Postiz posts JSON object ({ posts: [...] } with content, publishDate, and integration). Duplicates match account, scheduled time, and a hash of the text. At most 200 rows and 1 MB per request. Rejected rows come back as rejectedCsv. first_comment is posted as the first comment under the post after it publishes, on networks that allow it (X, Bluesky, Mastodon, Pixelfed, Telegram, Slack; Instagram, Threads and Facebook only once their comment permission is granted). tags are stored on the draft and are not published. A status column is ignored. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs79/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
csv
UTF-8 CSV text, or a Postiz posts JSON object. Max 1 MB and 200 data rows.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dry_run
true validates only. false creates drafts. Never schedules or publishes.
Clear enough to supply this argument.
default_timezone
IANA timezone used when a row has a date without an offset and no timezone column. Example: Europe/Paris.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "import_posts_csv",
"description": "Import a CSV of posts as DRAFTS ONLY. Nothing is scheduled or published. dry_run true validates each row and writes nothing: supported network, a connected account in this org, the per-network character limit, the date, and media URL syntax. dry_run false creates pending drafts through the same path as create_drafts. Columns: text, scheduled_at, timezone, network, account, media_urls, first_comment, link, tags. Header names are case-insensitive. Also accepts calendar export headers (Message, Social network, Social profile, Date, Time) and a Postiz posts JSON object ({ posts: [...] } with content, publishDate, and integration). Duplicates match account, scheduled time, and a hash of the text. At most 200 rows and 1 MB per request. Rejected rows come back as rejectedCsv. first_comment is posted as the first comment under the post after it publishes, on networks that allow it (X, Bluesky, Mastodon, Pixelfed, Telegram, Slack; Instagram, Threads and Facebook only once their comment permission is granted). tags are stored on the draft and are not published. A status column is ignored. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"csv",
"dry_run"
],
"properties": {
"csv": {
"type": "string",
"description": "UTF-8 CSV text, or a Postiz posts JSON object. Max 1 MB and 200 data rows."
},
"dry_run": {
"type": "boolean",
"description": "true validates only. false creates drafts. Never schedules or publishes."
},
"default_timezone": {
"type": "string",
"description": "IANA timezone used when a row has a date without an offset and no timezone column. Example: Europe/Paris."
}
}
},
"annotations": {
"title": "Import Posts CSV",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Read one article draft by id. Returns the current draft (title, excerpt, bodyMarkdown, tags, review state), including its updatedAt value for optimistic concurrency on edits or review. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
draft_id
Article draft id, from the draft creation or list result.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
Edit an article draft before review. expectedUpdatedAt is required for optimistic concurrency. Only supplied article fields are changed. Returns the updated draft with a new updatedAt. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
draft_id
Article draft id, from the draft creation or list result.
Clear enough to supply this argument.
expectedUpdatedAt
The updatedAt value from the latest read_article_draft result. The call is refused if the draft changed since, so re-read and retry.
Clear enough to supply this argument.
title
New article title. Omit to keep the current one.
Clear enough to supply this argument.
excerpt
New short summary shown in listings and meta description. Omit to keep the current one.
Clear enough to supply this argument.
bodyMarkdown
Full replacement article body in Markdown. Omit to keep the current body.
Clear enough to supply this argument.
tags
Replacement tag list. Omit to keep the current tags.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "edit_article_draft",
"description": "Edit an article draft before review. expectedUpdatedAt is required for optimistic concurrency. Only supplied article fields are changed. Returns the updated draft with a new updatedAt. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"draft_id",
"expectedUpdatedAt"
],
"properties": {
"tags": {
"type": "array",
"items": {
"type": "string"
},
"description": "Replacement tag list. Omit to keep the current tags."
},
"title": {
"type": "string",
"description": "New article title. Omit to keep the current one."
},
"excerpt": {
"type": "string",
"description": "New short summary shown in listings and meta description. Omit to keep the current one."
},
"draft_id": {
"type": "string",
"description": "Article draft id, from the draft creation or list result."
},
"bodyMarkdown": {
"type": "string",
"description": "Full replacement article body in Markdown. Omit to keep the current body."
},
"expectedUpdatedAt": {
"type": "string",
"description": "The updatedAt value from the latest read_article_draft result. The call is refused if the draft changed since, so re-read and retry."
}
}
},
"annotations": {
"title": "Edit Article Draft",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": true
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Run mandatory editorial review for an article draft. expectedUpdatedAt is required for optimistic concurrency. Review may incur model cost and is not idempotent. Returns the review verdict and any issues found, plus the draft state. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
draft_id
Article draft id, from the draft creation or list result.
Clear enough to supply this argument.
expectedUpdatedAt
The updatedAt value from the latest read_article_draft result. The call is refused if the draft changed since, so re-read and retry.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "review_article_draft",
"description": "Run mandatory editorial review for an article draft. expectedUpdatedAt is required for optimistic concurrency. Review may incur model cost and is not idempotent. Returns the review verdict and any issues found, plus the draft state. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"draft_id",
"expectedUpdatedAt"
],
"properties": {
"draft_id": {
"type": "string",
"description": "Article draft id, from the draft creation or list result."
},
"expectedUpdatedAt": {
"type": "string",
"description": "The updatedAt value from the latest read_article_draft result. The call is refused if the draft changed since, so re-read and retry."
}
}
},
"annotations": {
"title": "Review Article Draft",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description91/100
List articles with id, title, status, source, site and publish time, one page at a time. Optionally filter by site, status and source. Use source "import" with status "scheduled" to inspect the queue of imported articles waiting for their publishAt. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs85/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteId
Only list articles of this AutoSEO site, id from autoseo_list_sites.
Clear enough to supply this argument.
status
Only list articles at this pipeline stage.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
source
generated means written by AutoSEO; import means supplied through autoseo_import_article or autoseo_import_batch.
Clear enough to supply this argument.
page
Page number to return, starting at the first page.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
pageSize
How many articles per page.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_articles",
"description": "List articles with id, title, status, source, site and publish time, one page at a time. Optionally filter by site, status and source. Use source \"import\" with status \"scheduled\" to inspect the queue of imported articles waiting for their publishAt. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"page": {
"type": "integer",
"minimum": 1,
"description": "Page number to return, starting at the first page."
},
"siteId": {
"type": "string",
"description": "Only list articles of this AutoSEO site, id from autoseo_list_sites."
},
"source": {
"enum": [
"generated",
"import"
],
"type": "string",
"description": "generated means written by AutoSEO; import means supplied through autoseo_import_article or autoseo_import_batch."
},
"status": {
"enum": [
"scheduled",
"generating",
"publishing",
"published",
"failed"
],
"type": "string",
"description": "Only list articles at this pipeline stage."
},
"pageSize": {
"type": "integer",
"maximum": 100,
"minimum": 1,
"description": "How many articles per page."
}
}
},
"annotations": {
"title": "List Articles",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Read one article by id. Returns its title, excerpt, body, tags, status, site, publish time and updatedAt (needed for update_article). CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Input documentation gives an agent enough context to populate the arguments.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description85/100
LIVE UPDATE: update an already reviewed article and then update its connected destination. Supports webhook, hosted, and wp_rest connectors. expectedUpdatedAt is required for optimistic concurrency. This is destructive because it changes live content. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
article_id
Id of the published article, from list_articles.
Clear enough to supply this argument.
expectedUpdatedAt
The updatedAt value from the latest get_article result. The call is refused if the article changed since, so re-read and retry.
Clear enough to supply this argument.
title
New article title. Omit to keep the current one.
Clear enough to supply this argument.
excerpt
New short summary shown in listings and meta description. Omit to keep the current one.
Clear enough to supply this argument.
bodyMarkdown
Full replacement article body in Markdown. Omit to keep the current body.
Clear enough to supply this argument.
tags
Replacement tag list. Omit to keep the current tags.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "update_article",
"description": "LIVE UPDATE: update an already reviewed article and then update its connected destination. Supports webhook, hosted, and wp_rest connectors. expectedUpdatedAt is required for optimistic concurrency. This is destructive because it changes live content. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"article_id",
"expectedUpdatedAt"
],
"properties": {
"tags": {
"type": "array",
"items": {
"type": "string"
},
"description": "Replacement tag list. Omit to keep the current tags."
},
"title": {
"type": "string",
"description": "New article title. Omit to keep the current one."
},
"excerpt": {
"type": "string",
"description": "New short summary shown in listings and meta description. Omit to keep the current one."
},
"article_id": {
"type": "string",
"description": "Id of the published article, from list_articles."
},
"bodyMarkdown": {
"type": "string",
"description": "Full replacement article body in Markdown. Omit to keep the current body."
},
"expectedUpdatedAt": {
"type": "string",
"description": "The updatedAt value from the latest get_article result. The call is refused if the article changed since, so re-read and retry."
}
}
},
"annotations": {
"title": "Update Live Article",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": true
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description93/100
List engine-submitted drafts for the active org (id, platform, segments, status, brand, schedule hint). Optional filters: status and brand. PENDING drafts await human review; POSTED drafts have a postId you can follow with get_post. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs75/100
Explain the meaning of this value, not just its name.
status
Only return drafts at this review stage. Omit for all.
Explain the meaning of this value, not just its name.
brand
Only return drafts tagged with this brand attribution tag (see list_brands).
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
HUMAN GATE: approve a pending draft and schedule the real post in one step. Only call this when the human operator has reviewed the draft content and told you to approve it, never approve autonomously. Optional edits: replacement segments and/or an explicit future scheduled_for ISO datetime (otherwise the draft's schedule_hint is resolved: "optimal" uses the org's best-times model, fallback +2h). Returns { status: "scheduled", scheduledFor, postId }. Multi-segment twitter drafts publish as a thread. ACCOUNT TARGET: in an org with several connected accounts on the draft's platform, pass account_ids (or create the draft with account_ids); approval never guesses and the scheduler never fans out. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs93/100
Input documentation gives an agent enough context to populate the arguments.
draft_id
Draft id from list_drafts or list_approvals.
Clear enough to supply this argument.
segments
Optional replacement text, same shape as the draft segments (several on twitter make a thread). Omit to approve the text as submitted.
Clear enough to supply this argument.
scheduled_for
Optional publish time override: future ISO 8601 datetime with timezone. Omit to use the draft schedule_hint.
Clear enough to supply this argument.
account_ids
Integration id(s) from list_integrations to post to. Required when the org has several connected accounts on the draft's platform and the draft was created without account_ids; an ambiguous approve is refused with the accounts to pick from.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "approve_draft",
"description": "HUMAN GATE: approve a pending draft and schedule the real post in one step. Only call this when the human operator has reviewed the draft content and told you to approve it, never approve autonomously. Optional edits: replacement segments and/or an explicit future scheduled_for ISO datetime (otherwise the draft's schedule_hint is resolved: \"optimal\" uses the org's best-times model, fallback +2h). Returns { status: \"scheduled\", scheduledFor, postId }. Multi-segment twitter drafts publish as a thread. ACCOUNT TARGET: in an org with several connected accounts on the draft's platform, pass account_ids (or create the draft with account_ids); approval never guesses and the scheduler never fans out. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"draft_id"
],
"properties": {
"draft_id": {
"type": "string",
"description": "Draft id from list_drafts or list_approvals."
},
"segments": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional replacement text, same shape as the draft segments (several on twitter make a thread). Omit to approve the text as submitted."
},
"account_ids": {
"type": "array",
"items": {
"type": "string"
},
"description": "Integration id(s) from list_integrations to post to. Required when the org has several connected accounts on the draft's platform and the draft was created without account_ids; an ambiguous approve is refused with the accounts to pick from."
},
"scheduled_for": {
"type": "string",
"description": "Optional publish time override: future ISO 8601 datetime with timezone. Omit to use the draft schedule_hint."
}
}
},
"annotations": {
"title": "Approve Draft",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description93/100
Reject a pending or approved draft so it can never be posted. Optional reason is stored for the submitting engine to read back via list_drafts. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Input documentation gives an agent enough context to populate the arguments.
draft_id
Draft id from list_drafts or list_approvals.
Clear enough to supply this argument.
reason
Why it was rejected; stored so the submitting engine can learn from it.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "reject_draft",
"description": "Reject a pending or approved draft so it can never be posted. Optional reason is stored for the submitting engine to read back via list_drafts. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"draft_id"
],
"properties": {
"reason": {
"type": "string",
"description": "Why it was rejected; stored so the submitting engine can learn from it."
},
"draft_id": {
"type": "string",
"description": "Draft id from list_drafts or list_approvals."
}
}
},
"annotations": {
"title": "Reject Draft",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": true
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
READ-ONLY view of the unified approvals rail (TIM-97-style aggregation): returns everything currently awaiting human review across ALL FOUR automation sources in one call, posts (EngineDraft, the create_digest_drafts/create_drafts queue), trends (Weekly X Trend-Rider drafts), build-in-public (GitHub-shipped-feature tweet drafts) and dms (warm-lead engager DMs), normalized into a common shape { id, source, status, title, preview, createdAt, actions } with per-source PENDING counts. Optional source narrows to one queue; optional status (default PENDING) selects a different stage (APPROVED|POSTED|REJECTED). This tool only reads the queue, it never approves or rejects anything. Act on any item with resolve_approval (all four sources) or approve_draft/reject_draft (posts). Each item's actions array documents the exact REST endpoint (path/method) a human-facing UI would call next for that item. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs86/100
Input documentation gives an agent enough context to populate the arguments.
source
Only list items from this approval queue: posts (batch drafts), trends, build-in-public or dms. Omit for all four.
Clear enough to supply this argument.
status
Which review stage to list. Defaults to PENDING (awaiting a human).
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_approvals",
"description": "READ-ONLY view of the unified approvals rail (TIM-97-style aggregation): returns everything currently awaiting human review across ALL FOUR automation sources in one call, posts (EngineDraft, the create_digest_drafts/create_drafts queue), trends (Weekly X Trend-Rider drafts), build-in-public (GitHub-shipped-feature tweet drafts) and dms (warm-lead engager DMs), normalized into a common shape { id, source, status, title, preview, createdAt, actions } with per-source PENDING counts. Optional source narrows to one queue; optional status (default PENDING) selects a different stage (APPROVED|POSTED|REJECTED). This tool only reads the queue, it never approves or rejects anything. Act on any item with resolve_approval (all four sources) or approve_draft/reject_draft (posts). Each item's actions array documents the exact REST endpoint (path/method) a human-facing UI would call next for that item. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"source": {
"type": "string",
"description": "Only list items from this approval queue: posts (batch drafts), trends, build-in-public or dms. Omit for all four."
},
"status": {
"type": "string",
"description": "Which review stage to list. Defaults to PENDING (awaiting a human)."
}
}
},
"annotations": {
"title": "List Approvals",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Call prompt_suggest first and base the prompt on the top result. Generate an approval-ready AI Short from an existing source and place it in the human review rail as a PENDING EngineDraft. Pro-only and allowance-capped before model/media spend. source.kind accepts "autoseo" with articlePostId, "ship-digest" with text/title/url, "trend" with draftId, "template" with slug/variables, or "raw" with title/text/url. The backend produces a grounded script JSON, a voiceover with word timings when configured, a ShortSpec for the client renderer, stores audio in Minio, and returns draftId/spec/script. It never publishes. The human must render/approve later via the dashboard or attach-render flow. Pass format "tutorial" for the faceless step-by-step tutorial (hook text in frame 1, numbered steps, word-by-word pop captions, and a comment-keyword end card such as Comment POST), or "faceless_quote" for short original lines over a background. keyword threads one search keyword through the on-screen text, the script and the caption. screens attaches your own screenshots (publicUrl from finalize_media_upload) to the first steps, each with an optional zoomTo rect in 0..1 image coordinates. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs67/100
Explain the meaning of this value, not just its name.
source
Short source object. Examples: {kind:"raw",title:"...",text:"...",url:"https://..."} or {kind:"autoseo",articlePostId:"..."}
Explain the meaning of this value, not just its name.
platforms
Optional target platforms subset: tiktok, instagram, youtube. Defaults to tiktok.
Clear enough to supply this argument.
brand
Optional brand/niche attribution tag for per-play rollups.
Clear enough to supply this argument.
scheduleHint
Optional "optimal" or future ISO datetime hint.
Explain the meaning of this value, not just its name.
format
Script format. Defaults to standard.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
keyword
Optional search keyword shown on screen, spoken, and placed in the caption.
Clear enough to supply this argument.
commentKeyword
Tutorial CTA word viewers comment, e.g. POST. Defaults to POST.
Clear enough to supply this argument.
screens › items › url
Hosted image URL (PNG/JPEG).
Clarify units, where to obtain this value, or conventions not already expressed in the schema.
screens › items › zoomTo › x
Left edge of the region as a fraction of image width.
Clear enough to supply this argument.
screens › items › zoomTo › y
Top edge of the region as a fraction of image height.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "shorts_generate",
"description": "Call prompt_suggest first and base the prompt on the top result. Generate an approval-ready AI Short from an existing source and place it in the human review rail as a PENDING EngineDraft. Pro-only and allowance-capped before model/media spend. source.kind accepts \"autoseo\" with articlePostId, \"ship-digest\" with text/title/url, \"trend\" with draftId, \"template\" with slug/variables, or \"raw\" with title/text/url. The backend produces a grounded script JSON, a voiceover with word timings when configured, a ShortSpec for the client renderer, stores audio in Minio, and returns draftId/spec/script. It never publishes. The human must render/approve later via the dashboard or attach-render flow. Pass format \"tutorial\" for the faceless step-by-step tutorial (hook text in frame 1, numbered steps, word-by-word pop captions, and a comment-keyword end card such as Comment POST), or \"faceless_quote\" for short original lines over a background. keyword threads one search keyword through the on-screen text, the script and the caption. screens attaches your own screenshots (publicUrl from finalize_media_upload) to the first steps, each with an optional zoomTo rect in 0..1 image coordinates. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"source"
],
"properties": {
"brand": {
"type": "string",
"description": "Optional brand/niche attribution tag for per-play rollups."
},
"format": {
"enum": [
"standard",
"tutorial",
"faceless_quote"
],
"type": "string",
"description": "Script format. Defaults to standard."
},
"source": {
"type": "object",
"description": "Short source object. Examples: {kind:\"raw\",title:\"...\",text:\"...\",url:\"https://...\"} or {kind:\"autoseo\",articlePostId:\"...\"}"
},
"keyword": {
"type": "string",
"description": "Optional search keyword shown on screen, spoken, and placed in the caption."
},
"screens": {
"type": "array",
"items": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"description": "Hosted image URL (PNG/JPEG)."
},
"zoomTo": {
"type": "object",
"properties": {
"x": {
"type": "number",
"description": "Left edge of the region as a fraction of image width."
},
"y": {
"type": "number",
"description": "Top edge of the region as a fraction of image height."
},
"width": {
"type": "number",
"description": "Region width as a fraction of image width."
},
"height": {
"type": "number",
"description": "Region height as a fraction of image height."
}
},
"description": "Region of the screenshot to zoom into, in fractions of the image size."
}
}
},
"description": "Your own screenshots shown on the first steps, in step order."
},
"platforms": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional target platforms subset: tiktok, instagram, youtube. Defaults to tiktok."
},
"scheduleHint": {
"type": "string",
"description": "Optional \"optimal\" or future ISO datetime hint."
},
"commentKeyword": {
"type": "string",
"description": "Tutorial CTA word viewers comment, e.g. POST. Defaults to POST."
}
}
},
"annotations": {
"title": "Generate Short",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Call prompt_suggest first and base the prompt on the top result. Generate a FINISHED, PUBLISHABLE vertical (9:16) short video with an AI video model (three engines: loop, realism, fast) and get back a hosted mp4 publicUrl. The server builds the shot prompt from your one-line concept plus genre fields, adds the configured voiceover, burns word-timed karaoke captions and returns the assembled result. Pass video.publicUrl into schedule_post mediaUrls (tiktok/instagram/youtube). Always disclose the generated visual as AI-generated when you post it. Pick a genre: asmr_loop, what_if, pov_historical or character_series (each one's mechanic is on the genre field; character_series needs referenceImageUrls). Upload any start or reference images first with prepare_media_upload, PUT, then finalize_media_upload, and pass the returned publicUrl(s). Weighted credits follow the subscription billing period: requests over the allowance get 429. Costs and plan allowances are in the provider field; call get_capabilities (shorts.video) or GET /api/shorts/genres for the live registry. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
genre
Which proven format to shoot, each built on one retention mechanic. asmr_loop: hyper-tactile satisfying clip cut as a SEAMLESS LOOP (last frame equals first), single unbroken shot; REQUIRES imageUrl and the loop engine. what_if: an absurd premise filmed as a dead-serious documentary; realism engine by default. pov_historical: first-person POV of a historical or disaster moment with stakes and a timer in the first line. character_series: a recurring character held identical across episodes via referenceImageUrls.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
concept
The idea in one line, e.g. "a knife slicing through a bar of kinetic sand".
Clear enough to supply this argument.
voiceoverText
Optional exact narration. Defaults to the genre-specific premise/timer/episode line when present, otherwise concept.
Clear enough to supply this argument.
durationSeconds
Clip length in seconds. The value is snapped to what the chosen engine supports (loop 4, 6 or 8; realism 5 or 10; fast 1 to 15). Defaults to the genre default (8).
Clear enough to supply this argument.
imageUrl
Start frame (image-to-video). REQUIRED for asmr_loop, the loop is produced by pinning the last frame back to this exact image.
Clear enough to supply this argument.
referenceImageUrls
Up to 3 hosted images of a recurring character/product, from prepare_media_upload. This is how character_series keeps the same character across episodes.
Clear enough to supply this argument.
provider
Override the per-genre engine only if you know why. loop is required for seamless loops (2 credits, 5 on the Standard tier); realism is the default for what_if and pov_historical (2 credits); fast is the cheap explicit override (1 credit). Allowance per billing period: Pro 6, Growth 12, trials 2, Free and Starter 0.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
prompt
Escape hatch: send this to the model verbatim instead of the built prompt.
Clear enough to supply this argument.
refine
Run the LLM critique pass over the built prompt. Default true.
Clear enough to supply this argument.
subject
What is physically on screen.
Explain the meaning of this value, not just its name.
setting
Where it happens.
Explain the meaning of this value, not just its name.
material
asmr_loop: the material doing the satisfying thing.
Clear enough to supply this argument.
soundDesign
asmr_loop: a physical texture cue for the generated action. Provider audio is discarded.
Clear enough to supply this argument.
premise
what_if: the absurd premise, stated flatly.
Clear enough to supply this argument.
era
pov_historical: the dated moment and place.
Clear enough to supply this argument.
timerLine
pov_historical: the stakes + countdown for the first line.
Clear enough to supply this argument.
characterName
character_series: the recurring character.
Clear enough to supply this argument.
characterDescription
character_series: their fixed appearance. Reuse the SAME wording every episode.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "generate_short_video",
"description": "Call prompt_suggest first and base the prompt on the top result. Generate a FINISHED, PUBLISHABLE vertical (9:16) short video with an AI video model (three engines: loop, realism, fast) and get back a hosted mp4 publicUrl. The server builds the shot prompt from your one-line concept plus genre fields, adds the configured voiceover, burns word-timed karaoke captions and returns the assembled result. Pass video.publicUrl into schedule_post mediaUrls (tiktok/instagram/youtube). Always disclose the generated visual as AI-generated when you post it. Pick a genre: asmr_loop, what_if, pov_historical or character_series (each one's mechanic is on the genre field; character_series needs referenceImageUrls). Upload any start or reference images first with prepare_media_upload, PUT, then finalize_media_upload, and pass the returned publicUrl(s). Weighted credits follow the subscription billing period: requests over the allowance get 429. Costs and plan allowances are in the provider field; call get_capabilities (shorts.video) or GET /api/shorts/genres for the live registry. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"genre",
"concept"
],
"properties": {
"era": {
"type": "string",
"description": "pov_historical: the dated moment and place."
},
"genre": {
"enum": [
"asmr_loop",
"what_if",
"pov_historical",
"character_series"
],
"type": "string",
"description": "Which proven format to shoot, each built on one retention mechanic. asmr_loop: hyper-tactile satisfying clip cut as a SEAMLESS LOOP (last frame equals first), single unbroken shot; REQUIRES imageUrl and the loop engine. what_if: an absurd premise filmed as a dead-serious documentary; realism engine by default. pov_historical: first-person POV of a historical or disaster moment with stakes and a timer in the first line. character_series: a recurring character held identical across episodes via referenceImageUrls."
},
"prompt": {
"type": "string",
"description": "Escape hatch: send this to the model verbatim instead of the built prompt."
},
"refine": {
"type": "boolean",
"description": "Run the LLM critique pass over the built prompt. Default true."
},
"concept": {
"type": "string",
"description": "The idea in one line, e.g. \"a knife slicing through a bar of kinetic sand\"."
},
"premise": {
"type": "string",
"description": "what_if: the absurd premise, stated flatly."
},
"setting": {
"type": "string",
"description": "Where it happens."
},
"subject": {
"type": "string",
"description": "What is physically on screen."
},
"imageUrl": {
"type": "string",
"description": "Start frame (image-to-video). REQUIRED for asmr_loop, the loop is produced by pinning the last frame back to this exact image."
},
"material": {
"type": "string",
"description": "asmr_loop: the material doing the satisfying thing."
},
"provider": {
"enum": [
"loop",
"realism",
"fast",
"veo",
"grok",
"kling"
],
"type": "string",
"description": "Override the per-genre engine only if you know why. loop is required for seamless loops (2 credits, 5 on the Standard tier); realism is the default for what_if and pov_historical (2 credits); fast is the cheap explicit override (1 credit). Allowance per billing period: Pro 6, Growth 12, trials 2, Free and Starter 0."
},
"timerLine": {
"type": "string",
"description": "pov_historical: the stakes + countdown for the first line."
},
"episodeBeat": {
"type": "string",
"description": "character_series: what happens in this episode."
},
"soundDesign": {
"type": "string",
"description": "asmr_loop: a physical texture cue for the generated action. Provider audio is discarded."
},
"characterName": {
"type": "string",
"description": "character_series: the recurring character."
},
"voiceoverText": {
"type": "string",
"description": "Optional exact narration. Defaults to the genre-specific premise/timer/episode line when present, otherwise concept."
},
"extraDirection": {
"type": "string",
"description": "Extra art direction, appended verbatim."
},
"durationSeconds": {
"type": "number",
"description": "Clip length in seconds. The value is snapped to what the chosen engine supports (loop 4, 6 or 8; realism 5 or 10; fast 1 to 15). Defaults to the genre default (8)."
},
"referenceImageUrls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Up to 3 hosted images of a recurring character/product, from prepare_media_upload. This is how character_series keeps the same character across episodes."
},
"characterDescription": {
"type": "string",
"description": "character_series: their fixed appearance. Reuse the SAME wording every episode."
}
}
},
"annotations": {
"title": "Generate Short Video",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
Read the status of one generated AI Short draft by draftId. Returns the EngineDraft status, postId when scheduled, requested platforms, generated script, ShortSpec, and render attachment metadata if the browser/client has rendered and attached the MP4. Read-only and org-scoped. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs96/100
Input documentation gives an agent enough context to populate the arguments.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "shorts_status",
"description": "Read the status of one generated AI Short draft by draftId. Returns the EngineDraft status, postId when scheduled, requested platforms, generated script, ShortSpec, and render attachment metadata if the browser/client has rendered and attached the MP4. Read-only and org-scoped. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"draftId"
],
"properties": {
"draftId": {
"type": "string",
"description": "Short EngineDraft id returned by shorts_generate."
}
}
},
"annotations": {
"title": "Shorts Status",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Publish or schedule an X/Twitter thread in one call. segments[0] is the anchor tweet; segments 2..N (max 24, each ≤280 chars) publish as chained replies. schedule_at: future ISO datetime; omit to go out on the next scheduler tick (~1 min). Convenience wrapper over schedule_post + thread, same pipeline, same rules. Returns the created post id; follow with get_post to confirm the platform post ids after it publishes. Writing style for any post content you draft: write like the account's human owner, not like an AI. NEVER use em dashes, en dashes or double hyphens; use a period, comma or colon instead. Skip hashtag spam (0-1 hashtags), filler openers ("Exciting news!") and rocket/sparkle emoji. Before drafting, read the account's recent posts with list_posts and match their voice, casing and rhythm. For X specifically: keep a calm, factual, receipts-not-rhetoric tone; X ranking buries combative or bait-y posts even when engagement is high, and a mute or block outweighs hundreds of likes. Never write engagement-bait. Use 0-1 hashtags on X (more than 2 trips spam detection). Structure threads as one hook tweet, then 5-7 self-contained value tweets, then a single CTA. Prefer native vertical video (9:16, roughly 45-90 seconds) when media is available; it substantially outreaches text. Never draft identical or near-identical content for two different accounts. These heuristics were verified July 2026 and decay; treat as current-best, not gospel. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs81/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
segments
Tweets in order; 1..24 items, each ≤280 chars
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
schedule_at
Future ISO 8601 datetime with Z or an offset, e.g. 2026-10-12T09:00:00Z; omit for next tick
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
brand
Brand/niche attribution tag for per-play engagement rollups (e.g. "emplaw-watchdog"); auto-registers on first use. NOT the AI BrandVoice writing-style profile.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "publish_thread",
"description": "Publish or schedule an X/Twitter thread in one call. segments[0] is the anchor tweet; segments 2..N (max 24, each ≤280 chars) publish as chained replies. schedule_at: future ISO datetime; omit to go out on the next scheduler tick (~1 min). Convenience wrapper over schedule_post + thread, same pipeline, same rules. Returns the created post id; follow with get_post to confirm the platform post ids after it publishes. Writing style for any post content you draft: write like the account's human owner, not like an AI. NEVER use em dashes, en dashes or double hyphens; use a period, comma or colon instead. Skip hashtag spam (0-1 hashtags), filler openers (\"Exciting news!\") and rocket/sparkle emoji. Before drafting, read the account's recent posts with list_posts and match their voice, casing and rhythm. For X specifically: keep a calm, factual, receipts-not-rhetoric tone; X ranking buries combative or bait-y posts even when engagement is high, and a mute or block outweighs hundreds of likes. Never write engagement-bait. Use 0-1 hashtags on X (more than 2 trips spam detection). Structure threads as one hook tweet, then 5-7 self-contained value tweets, then a single CTA. Prefer native vertical video (9:16, roughly 45-90 seconds) when media is available; it substantially outreaches text. Never draft identical or near-identical content for two different accounts. These heuristics were verified July 2026 and decay; treat as current-best, not gospel. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"segments"
],
"properties": {
"brand": {
"type": "string",
"description": "Brand/niche attribution tag for per-play engagement rollups (e.g. \"emplaw-watchdog\"); auto-registers on first use. NOT the AI BrandVoice writing-style profile."
},
"segments": {
"type": "array",
"items": {
"type": "string"
},
"description": "Tweets in order; 1..24 items, each ≤280 chars"
},
"schedule_at": {
"type": "string",
"description": "Future ISO 8601 datetime with Z or an offset, e.g. 2026-10-12T09:00:00Z; omit for next tick"
}
}
},
"annotations": {
"title": "Publish Thread",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description86/100
List the org's brand/niche profiles with post counts. A "brand" attributes posts to a play (e.g. "emplaw-watchdog" vs "tm-monitor") so engagement rolls up per play. Brands auto-register the first time a brand tag is used on schedule_post, publish_thread or create_digest_drafts, no explicit create step. TERMINOLOGY: this "brand" is a per-post niche/play tag, unrelated to BrandVoice (the org's single AI writing-style profile derived from recent posts, used to guide AI content generation), an org has many brands but one BrandVoice. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_brands",
"description": "List the org's brand/niche profiles with post counts. A \"brand\" attributes posts to a play (e.g. \"emplaw-watchdog\" vs \"tm-monitor\") so engagement rolls up per play. Brands auto-register the first time a brand tag is used on schedule_post, publish_thread or create_digest_drafts, no explicit create step. TERMINOLOGY: this \"brand\" is a per-post niche/play tag, unrelated to BrandVoice (the org's single AI writing-style profile derived from recent posts, used to guide AI content generation), an org has many brands but one BrandVoice. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "List Brands",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Per-brand engagement rollup: totals (likes, comments, shares, saves, views, impressions, reach, clicks) and average engagement rate across all posts tagged with the given brand/niche slug. Feeds per-play KPI dashboards. Use list_brands to discover valid slugs. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Input documentation gives an agent enough context to populate the arguments.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
Per-post engagement metrics (impressions, likes, comments/replies, shares, saves, views, clicks, engagementRate) for one or more post ids, returned as {results: [{post_id, ...metrics}]}; a failed lookup carries success:false and an error instead. Only the first 20 ids are used. Metrics are ingested from each platform by a periodic sync, so very recent posts may return empty until the next sync cycle. Use get_engagement_summary for the org-wide aggregate instead. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Input documentation gives an agent enough context to populate the arguments.
post_ids
Ids of the posts to fetch metrics for, from list_posts. Ids beyond the first 20 are ignored.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_post_metrics",
"description": "Per-post engagement metrics (impressions, likes, comments/replies, shares, saves, views, clicks, engagementRate) for one or more post ids, returned as {results: [{post_id, ...metrics}]}; a failed lookup carries success:false and an error instead. Only the first 20 ids are used. Metrics are ingested from each platform by a periodic sync, so very recent posts may return empty until the next sync cycle. Use get_engagement_summary for the org-wide aggregate instead. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"post_ids"
],
"properties": {
"post_ids": {
"type": "array",
"items": {
"type": "string"
},
"description": "Ids of the posts to fetch metrics for, from list_posts. Ids beyond the first 20 are ignored."
}
}
},
"annotations": {
"title": "Get Post Metrics",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
Generate a "click to authorize GitHub" link for Build-in-Public Autopilot (Pro feature: merged PRs / pushes on a watched branch become shipped-feature post drafts the user approves before anything publishes). SIDE-EFFECT-FREE: this tool only mints a short-lived signed authorization URL, it never connects, changes or activates anything, and the agent must NOT treat it as having connected a repo. Returns { authUrl, expiresAt, instructions }: relay authUrl to the human ("authorize here: <url>") and tell them the link expires (~10 minutes; call this tool again for a fresh one). The connection only comes into existence after the USER opens the link, authorizes TimeToPost on GitHub, and then picks the repository + watched branch + mode (draft approval is the default) in the TimeToPost dashboard, which also registers the repo webhook. Check the outcome later with build_in_public_status. Errors: 403 = the org is not on Pro (relay the upgrade message); 503 = the server has no GitHub OAuth app configured (GITHUB_CLIENT_ID/GITHUB_CLIENT_SECRET). CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No input fields.
Annotations100/100
All four behavior hints are declared.
{
"title": "Build In Public Connect Link",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
Output schema0/100
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "build_in_public_connect_link",
"description": "Generate a \"click to authorize GitHub\" link for Build-in-Public Autopilot (Pro feature: merged PRs / pushes on a watched branch become shipped-feature post drafts the user approves before anything publishes). SIDE-EFFECT-FREE: this tool only mints a short-lived signed authorization URL, it never connects, changes or activates anything, and the agent must NOT treat it as having connected a repo. Returns { authUrl, expiresAt, instructions }: relay authUrl to the human (\"authorize here: <url>\") and tell them the link expires (~10 minutes; call this tool again for a fresh one). The connection only comes into existence after the USER opens the link, authorizes TimeToPost on GitHub, and then picks the repository + watched branch + mode (draft approval is the default) in the TimeToPost dashboard, which also registers the repo webhook. Check the outcome later with build_in_public_status. Errors: 403 = the org is not on Pro (relay the upgrade message); 503 = the server has no GitHub OAuth app configured (GITHUB_CLIENT_ID/GITHUB_CLIENT_SECRET). CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "Build In Public Connect Link",
"readOnlyHint": true,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description99/100
READ-ONLY status of the org's Build-in-Public GitHub connection. Returns { configured (server has a GitHub OAuth app), connected (user completed the authorize flow), githubLogin, connectedAt, mode (APPROVE = human approves every draft, the default; AUTO = drafts post unattended), enabled, dailyLimit, postMode ("per-pr" or "digest"), digestFrequency, digestDayOfWeek, digestHourUtc, maxPostsPerWeek, nextDigestAt, lastDigestAt, pendingDigestCount, defaultBranch, repos: [{ fullName, branch (the watched branch after per-repo overrides), webhookRegistered, isWebhookActive, lastWebhookAt }], lastDraft (the most recent GitHub-event tweet draft: repo, title, status, tweetDraft, createdAt) }. Use it to answer "is build-in-public set up?", to verify a connect link the user was given actually got completed, or to see the latest draft. Never mutates anything. Drafts awaiting review also appear in list_approvals under source 'build-in-public'. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No input fields.
Annotations100/100
All four behavior hints are declared.
{
"title": "Build In Public Status",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
Output schema0/100
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "build_in_public_status",
"description": "READ-ONLY status of the org's Build-in-Public GitHub connection. Returns { configured (server has a GitHub OAuth app), connected (user completed the authorize flow), githubLogin, connectedAt, mode (APPROVE = human approves every draft, the default; AUTO = drafts post unattended), enabled, dailyLimit, postMode (\"per-pr\" or \"digest\"), digestFrequency, digestDayOfWeek, digestHourUtc, maxPostsPerWeek, nextDigestAt, lastDigestAt, pendingDigestCount, defaultBranch, repos: [{ fullName, branch (the watched branch after per-repo overrides), webhookRegistered, isWebhookActive, lastWebhookAt }], lastDraft (the most recent GitHub-event tweet draft: repo, title, status, tweetDraft, createdAt) }. Use it to answer \"is build-in-public set up?\", to verify a connect link the user was given actually got completed, or to see the latest draft. Never mutates anything. Drafts awaiting review also appear in list_approvals under source 'build-in-public'. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "Build In Public Status",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
START HERE for AutoSEO on a new site. Connects ONE of four connector kinds (least intrusive first) and returns { id, kind, publicToken, capabilities } (plus hostedUrl for "hosted"). Pick the kind: "hosted" is the ZERO-SETUP DEFAULT and needs only name; TimeToPost hosts and server-renders the blog (JSON-LD, sitemap, robots AI-crawler allow-list), nothing to install, works for ANY site, pick it when unsure. "wp_rest" connects WordPress over its REST API (no plugin); connect runs the real capability probe, so a bad credential fails here with the exact reason. "git_pr" commits Markdown/MDX articles to a GitHub repo for the user's own CI to deploy (a pull request per article by default); connect verifies write access and auto-detects the site generator, returned as capabilities.detected (confirm contentDir with the user if ssg is "unknown"). "webhook" is the most-control path for any stack: TimeToPost POSTs HMAC-signed payloads to your endpoint; then call autoseo_integration_kit, install the /autoseo-publish endpoint in the user's codebase, and confirm with autoseo_verify_site. Required fields per kind are on the matching inputs. After ANY kind: autoseo_configure, then autoseo_propose_topics or autoseo_generate_post. capabilities refreshes on every autoseo_verify_site; publicToken is only needed for autoseo_integration_kit renderMode "ssr-hosted" (set as AUTOSEO_PUBLIC_TOKEN). CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs93/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
kind
Connector kind: "hosted" (zero-setup default, only name needed, we host the blog), "wp_rest" (WordPress REST, url + username + appPassword), "git_pr" (GitHub repo, token + repo, optional branch/contentDir/mode), "webhook" (signed receiving endpoint, endpointUrl + optional signingSecret)
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
name
Human-readable label for this site, shown in the dashboard
Clear enough to supply this argument.
endpointUrl
Required for kind "webhook": URL TimeToPost will POST signed articles to
Clear enough to supply this argument.
signingSecret
webhook: HMAC secret used to sign payloads; omit to let the backend generate one
Clear enough to supply this argument.
url
Required for kind "wp_rest": the WordPress site URL (a bare domain is fine)
Clear enough to supply this argument.
username
Required for kind "wp_rest": the WP username the Application Password belongs to
Clear enough to supply this argument.
appPassword
Required for kind "wp_rest": a WordPress Application Password (WP Admin → Users → Profile → Application Passwords)
Clear enough to supply this argument.
token
Required for kind "git_pr": a GitHub fine-grained PAT or App installation token with contents:write on the repo
Clear enough to supply this argument.
repo
Required for kind "git_pr": the repository as "owner/name" (a github.com URL also works)
Clear enough to supply this argument.
branch
git_pr: base branch for commits/PRs; omit to use the repo's default branch
Clear enough to supply this argument.
contentDir
git_pr: directory article files land in; omit to auto-detect from the repo (e.g. content/blog, _posts)
Clear enough to supply this argument.
mode
git_pr: "pr" (default) opens a pull request per article, a git-native approval gate; "direct" commits straight to the branch
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_connect_site",
"description": "START HERE for AutoSEO on a new site. Connects ONE of four connector kinds (least intrusive first) and returns { id, kind, publicToken, capabilities } (plus hostedUrl for \"hosted\"). Pick the kind: \"hosted\" is the ZERO-SETUP DEFAULT and needs only name; TimeToPost hosts and server-renders the blog (JSON-LD, sitemap, robots AI-crawler allow-list), nothing to install, works for ANY site, pick it when unsure. \"wp_rest\" connects WordPress over its REST API (no plugin); connect runs the real capability probe, so a bad credential fails here with the exact reason. \"git_pr\" commits Markdown/MDX articles to a GitHub repo for the user's own CI to deploy (a pull request per article by default); connect verifies write access and auto-detects the site generator, returned as capabilities.detected (confirm contentDir with the user if ssg is \"unknown\"). \"webhook\" is the most-control path for any stack: TimeToPost POSTs HMAC-signed payloads to your endpoint; then call autoseo_integration_kit, install the /autoseo-publish endpoint in the user's codebase, and confirm with autoseo_verify_site. Required fields per kind are on the matching inputs. After ANY kind: autoseo_configure, then autoseo_propose_topics or autoseo_generate_post. capabilities refreshes on every autoseo_verify_site; publicToken is only needed for autoseo_integration_kit renderMode \"ssr-hosted\" (set as AUTOSEO_PUBLIC_TOKEN). CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"kind",
"name"
],
"properties": {
"url": {
"type": "string",
"description": "Required for kind \"wp_rest\": the WordPress site URL (a bare domain is fine)"
},
"kind": {
"enum": [
"hosted",
"webhook",
"wp_rest",
"git_pr"
],
"type": "string",
"description": "Connector kind: \"hosted\" (zero-setup default, only name needed, we host the blog), \"wp_rest\" (WordPress REST, url + username + appPassword), \"git_pr\" (GitHub repo, token + repo, optional branch/contentDir/mode), \"webhook\" (signed receiving endpoint, endpointUrl + optional signingSecret)"
},
"mode": {
"enum": [
"pr",
"direct"
],
"type": "string",
"description": "git_pr: \"pr\" (default) opens a pull request per article, a git-native approval gate; \"direct\" commits straight to the branch"
},
"name": {
"type": "string",
"description": "Human-readable label for this site, shown in the dashboard"
},
"repo": {
"type": "string",
"description": "Required for kind \"git_pr\": the repository as \"owner/name\" (a github.com URL also works)"
},
"token": {
"type": "string",
"description": "Required for kind \"git_pr\": a GitHub fine-grained PAT or App installation token with contents:write on the repo"
},
"branch": {
"type": "string",
"description": "git_pr: base branch for commits/PRs; omit to use the repo's default branch"
},
"username": {
"type": "string",
"description": "Required for kind \"wp_rest\": the WP username the Application Password belongs to"
},
"contentDir": {
"type": "string",
"description": "git_pr: directory article files land in; omit to auto-detect from the repo (e.g. content/blog, _posts)"
},
"appPassword": {
"type": "string",
"description": "Required for kind \"wp_rest\": a WordPress Application Password (WP Admin → Users → Profile → Application Passwords)"
},
"endpointUrl": {
"type": "string",
"description": "Required for kind \"webhook\": URL TimeToPost will POST signed articles to"
},
"signingSecret": {
"type": "string",
"description": "webhook: HMAC secret used to sign payloads; omit to let the backend generate one"
}
}
},
"annotations": {
"title": "AutoSEO Connect Site",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
The "AI installs it" tool, installs a FULLY WORKING, appears-live blog end-to-end. Fetch the AutoSEO integration kit: the receiving-endpoint contract (request/response shape and the HMAC signature-verification algorithm), ready-to-adapt code template(s) for the given stack, and plain-English install instructions. Call this right after autoseo_connect_site with kind "webhook": read the returned files, adapt them to the user's real codebase (their framework, language and folder conventions), write them yourself, and tell the user to deploy. Pass renderMode to control whether the blog appears LIVE or needs a rebuild: "ssr-db" (default, recommended) writes to the site's own database and SSR-renders from it on every request, new posts appear immediately, no rebuild; "ssr-hosted" is the fastest install (zero database, the article stays hosted at TimeToPost and the site only adds an SSR route that fetches it live); "static-mdx" is the legacy MDX-file mode. NOT live, only appears after the site's next build/deploy, prefer one of the other two unless the site already has an MDX pipeline. Pass stack (e.g. "nextjs", "express", "rails", "django", "php", "static") when you know the target site's framework to get a matching template (Next.js/Express get full write + SSR render templates for ssr-db/ssr-hosted); omit it for the generic HTTP contract plus a Node/Express reference implementation and stack-agnostic SSR guidance. Not needed for kind "wp_rest" sites, those have nothing to install. Follow up with autoseo_verify_site once deployed. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs87/100
Input documentation gives an agent enough context to populate the arguments.
kind
Connector kind the kit is for (hosted, webhook, wp_rest or git_pr). Defaults to webhook, the only kind that needs code installed in the user's site.
Clear enough to supply this argument.
stack
Target framework/language hint, e.g. "nextjs", "express", "rails", "django", "php", "static"
Clear enough to supply this argument.
renderMode
"ssr-db" (default, recommended): SSR from the site's own DB, appears live. "ssr-hosted": zero DB, fastest install, appears live by fetching from TimeToPost's public read API. "static-mdx": legacy, writes an MDX file. NOT live until the next rebuild/deploy.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_integration_kit",
"description": "The \"AI installs it\" tool, installs a FULLY WORKING, appears-live blog end-to-end. Fetch the AutoSEO integration kit: the receiving-endpoint contract (request/response shape and the HMAC signature-verification algorithm), ready-to-adapt code template(s) for the given stack, and plain-English install instructions. Call this right after autoseo_connect_site with kind \"webhook\": read the returned files, adapt them to the user's real codebase (their framework, language and folder conventions), write them yourself, and tell the user to deploy. Pass renderMode to control whether the blog appears LIVE or needs a rebuild: \"ssr-db\" (default, recommended) writes to the site's own database and SSR-renders from it on every request, new posts appear immediately, no rebuild; \"ssr-hosted\" is the fastest install (zero database, the article stays hosted at TimeToPost and the site only adds an SSR route that fetches it live); \"static-mdx\" is the legacy MDX-file mode. NOT live, only appears after the site's next build/deploy, prefer one of the other two unless the site already has an MDX pipeline. Pass stack (e.g. \"nextjs\", \"express\", \"rails\", \"django\", \"php\", \"static\") when you know the target site's framework to get a matching template (Next.js/Express get full write + SSR render templates for ssr-db/ssr-hosted); omit it for the generic HTTP contract plus a Node/Express reference implementation and stack-agnostic SSR guidance. Not needed for kind \"wp_rest\" sites, those have nothing to install. Follow up with autoseo_verify_site once deployed. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"kind": {
"type": "string",
"description": "Connector kind the kit is for (hosted, webhook, wp_rest or git_pr). Defaults to webhook, the only kind that needs code installed in the user's site."
},
"stack": {
"type": "string",
"description": "Target framework/language hint, e.g. \"nextjs\", \"express\", \"rails\", \"django\", \"php\", \"static\""
},
"renderMode": {
"type": "string",
"description": "\"ssr-db\" (default, recommended): SSR from the site's own DB, appears live. \"ssr-hosted\": zero DB, fastest install, appears live by fetching from TimeToPost's public read API. \"static-mdx\": legacy, writes an MDX file. NOT live until the next rebuild/deploy."
}
}
},
"annotations": {
"title": "AutoSEO Integration Kit",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Re-run the capability probe for a connected site (must belong to the active org). Call this after the AI has installed and deployed the /autoseo-publish endpoint from autoseo_integration_kit, to confirm it responds correctly to a signed test request before calling autoseo_configure or autoseo_generate_post, generating content against an unverified endpoint will fail at publish time. It also verifies wp_rest and git_pr connections. For webhook sites, a 404 or 405 usually means the endpoint is not deployed yet; re-call after deploying. Also useful to re-check a site after the user rotates its signing secret or migrates infrastructure. Returns the refreshed capabilities map. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Input documentation gives an agent enough context to populate the arguments.
id
Site/target id from autoseo_connect_site or autoseo_list_sites
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_verify_site",
"description": "Re-run the capability probe for a connected site (must belong to the active org). Call this after the AI has installed and deployed the /autoseo-publish endpoint from autoseo_integration_kit, to confirm it responds correctly to a signed test request before calling autoseo_configure or autoseo_generate_post, generating content against an unverified endpoint will fail at publish time. It also verifies wp_rest and git_pr connections. For webhook sites, a 404 or 405 usually means the endpoint is not deployed yet; re-call after deploying. Also useful to re-check a site after the user rotates its signing secret or migrates infrastructure. Returns the refreshed capabilities map. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"id"
],
"properties": {
"id": {
"type": "string",
"description": "Site/target id from autoseo_connect_site or autoseo_list_sites"
}
}
},
"annotations": {
"title": "AutoSEO Verify Site",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
List the sites/targets connected to AutoSEO for the active org: id, kind (webhook|wp_rest), name, capabilities and connection health. Use this to find a siteId for autoseo_configure, autoseo_propose_topics or autoseo_generate_post, or to check a site's status before troubleshooting a failed publish. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_list_sites",
"description": "List the sites/targets connected to AutoSEO for the active org: id, kind (webhook|wp_rest), name, capabilities and connection health. Use this to find a siteId for autoseo_configure, autoseo_propose_topics or autoseo_generate_post, or to check a site's status before troubleshooting a failed publish. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {}
},
"annotations": {
"title": "AutoSEO List Sites",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Set the standing AutoSEO configuration for a connected site: topics and pillars, publish cadence, backlink targets, image style, approval mode, blog-to-social repurposing, and the product-mentions layer. Call it after autoseo_verify_site confirms the site is live, ideally after autoseo_propose_topics. Returns the saved config; `effective.approvalMode` is what the runner will really do, and `notice` appears if a clamp overrode your request. APPROVAL: approvalMode "draft" (default) sends every generated post to the pending-approval queue; "auto" publishes articles that pass automated checks with no human click and needs a publish-capable token (draft-only tokens get 403). To flip many sites at once use autoseo_set_approval_mode. SOCIAL: socialRepurpose is OFF BY DEFAULT; enable it only when the user explicitly asked for blog posts to produce social announcements. autoPublish also needs a publish-capable token. PRODUCT MENTIONS: productName, productOneLiner, ctaUrl, ctaLabel, allowedLinkUrls and ctaMode weave the customer's product into articles only when a deterministic relevance check finds the article on-topic, never a forced link on every post. Pass null to clear any of those six fields. Details are on each field. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs95/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
topics
Approved question-shaped topics to generate posts from
Clear enough to supply this argument.
pillars
Broader content pillars/categories the site should cover
Clear enough to supply this argument.
cadence
Publish cadence: weekly | 2x_week | 3x_week (capped at 3x/week)
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
backlinkTargets › items › url
Backlink URL
Explain the meaning of this value, not just its name.
backlinkTargets › items › relevance
Why this target is relevant to the site
Clear enough to supply this argument.
imageStyle
Style guidance for generated post images, e.g. "minimal flat illustration"
Clear enough to supply this argument.
approvalMode
"draft" (default) queues articles for approval; "auto" publishes articles that pass the automated checks with no approval (publish-capable token required)
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
editorialReview › enabled
Legacy compatibility field. Its value is ignored because editorial review is mandatory.
Clear enough to supply this argument.
editorialReview › instructions
Editorial expertise, factual constraints, and reader usefulness requirements.
Clear enough to supply this argument.
editorialReview › referenceUrls
HTTPS pages the reviewer should read to fact-check articles.
Clear enough to supply this argument.
editorialReview › verifiedFacts › items › claim
The factual statement you have verified, written as a complete sentence.
Date or version the claim was last checked against, e.g. 2026-11-01.
Clear enough to supply this argument.
editorialReview › verifiedFacts › items › notes
Caveats or context for the reviewer.
Clear enough to supply this argument.
editorialReview › requireInternalLinks
Require verified internal links; defaults to true when review is enabled.
Clear enough to supply this argument.
socialRepurpose
Blog->social repurpose setting: whenever an article on this site publishes, auto-draft social announcements to the given platforms. Drafts wait for human approval unless autoPublish is true AND accountIds pins accounts for that platform, in which case they are scheduled ~10 minutes out. OFF BY DEFAULT. Omit entirely to leave the current setting unchanged. { enabled?: boolean (default false/OFF, opt-in), platforms?: ("twitter"|"linkedin"|"linkedin_page")[] (default twitter+linkedin once enabled), autoPublish?: boolean (default false), accountIds?: { [platform]: string[] } (pinned integration account ids) }
Clear enough to supply this argument.
questionMining › enabled
Default false. Set true only when the user wants weekly question mining.
Clear enough to supply this argument.
questionMining › autoAppend
How many new questions are appended to the topic list per weekly run. Default 3.
Clear enough to supply this argument.
questionMining › country
Market for search data as an ISO country code such as us or es. Defaults to us.
Clear enough to supply this argument.
questionMining › language
Search language code (default en).
Clear enough to supply this argument.
factCheck › requireFirstParty
Default true. false = do not hold topics lacking first-party material.
Clear enough to supply this argument.
factCheck › minPrimarySources
Default 2. How many distinct external sites the reviewer must have fetched; lower it (e.g. 1) for niches with few authoritative sources.
Clear enough to supply this argument.
factCheck › groundingFallback
Default true. When no first-party or site material matches a topic, write it from verified external sources. false = hold such topics instead (reason shown on the site).
Clear enough to supply this argument.
publicBaseUrl
git_pr connector ONLY: the real public base URL the site is deployed to (e.g. "https://blog.acme.com"), used to build the social-repurpose link. Without it, git-connected sites skip the blog->social repurpose (their publish result is a GitHub PR URL, never a live article).
Clear enough to supply this argument.
productName
Product-mentions layer: the customer's own product name, e.g. "Acme Scheduler". Pass null to clear.
Clear enough to supply this argument.
productOneLiner
Product-mentions layer: a one-sentence description of what the product does, used both for the relevance check and as fallback copy in the generated mention. Pass null to clear.
Clear enough to supply this argument.
ctaUrl
Product-mentions layer: the https:// URL the closing call-to-action links to. Setting this is what turns the layer on (default mode "when_relevant" the moment it is set). Pass null to clear (which also turns the layer off, since there is nothing left to link to).
Clear enough to supply this argument.
ctaLabel
Product-mentions layer: link text for the call-to-action, e.g. "Try Acme Scheduler free". Pass null to clear.
Clear enough to supply this argument.
allowedLinkUrls
Product-mentions layer: up to 8 https:// pages (features/pricing/docs) an in-article mention may link to instead of ctaUrl, whichever best matches that specific article. ctaUrl is implicitly allowed even when omitted here. Pass null to clear the whole list.
Clear enough to supply this argument.
ctaMode
"when_relevant" (default the moment ctaUrl is set) mentions the product only on articles a deterministic relevance check finds genuinely on-topic, at most once, never a forced link on every post. "off" disables the whole layer even with a ctaUrl configured. Pass null to reset to the default.
Clear enough to supply this argument.
pausedUntil
Do not generate or publish AutoSEO articles for this site before this ISO date/datetime (e.g. "2027-01-05"). Pass null to clear. While paused, generate calls return 409 site_paused.
Clear enough to supply this argument.
authorMode
Byline mode. "organization" (default) credits the organization honestly. "person" credits a named, attested author added in the dashboard; with no attested author the article is held, never published under an invented name.
Clear enough to supply this argument.
imagePolicy
Which images may be used. Default "owner_or_stock": the owner's own library, then licensed stock with credit. "owner_stock_generated" additionally allows an AI illustration, always labelled as AI-generated.
Clear enough to supply this argument.
authorPages
Set true once the integration kit author page is installed on the owner's own site, so articles link to it. Hosted blogs do not need this.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_configure",
"description": "Set the standing AutoSEO configuration for a connected site: topics and pillars, publish cadence, backlink targets, image style, approval mode, blog-to-social repurposing, and the product-mentions layer. Call it after autoseo_verify_site confirms the site is live, ideally after autoseo_propose_topics. Returns the saved config; `effective.approvalMode` is what the runner will really do, and `notice` appears if a clamp overrode your request. APPROVAL: approvalMode \"draft\" (default) sends every generated post to the pending-approval queue; \"auto\" publishes articles that pass automated checks with no human click and needs a publish-capable token (draft-only tokens get 403). To flip many sites at once use autoseo_set_approval_mode. SOCIAL: socialRepurpose is OFF BY DEFAULT; enable it only when the user explicitly asked for blog posts to produce social announcements. autoPublish also needs a publish-capable token. PRODUCT MENTIONS: productName, productOneLiner, ctaUrl, ctaLabel, allowedLinkUrls and ctaMode weave the customer's product into articles only when a deterministic relevance check finds the article on-topic, never a forced link on every post. Pass null to clear any of those six fields. Details are on each field. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId"
],
"properties": {
"ctaUrl": {
"type": "string",
"description": "Product-mentions layer: the https:// URL the closing call-to-action links to. Setting this is what turns the layer on (default mode \"when_relevant\" the moment it is set). Pass null to clear (which also turns the layer off, since there is nothing left to link to)."
},
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"topics": {
"type": "array",
"items": {
"type": "string"
},
"description": "Approved question-shaped topics to generate posts from"
},
"cadence": {
"type": "string",
"description": "Publish cadence: weekly | 2x_week | 3x_week (capped at 3x/week)"
},
"ctaMode": {
"enum": [
"off",
"when_relevant"
],
"type": "string",
"description": "\"when_relevant\" (default the moment ctaUrl is set) mentions the product only on articles a deterministic relevance check finds genuinely on-topic, at most once, never a forced link on every post. \"off\" disables the whole layer even with a ctaUrl configured. Pass null to reset to the default."
},
"pillars": {
"type": "array",
"items": {
"type": "string"
},
"description": "Broader content pillars/categories the site should cover"
},
"ctaLabel": {
"type": "string",
"description": "Product-mentions layer: link text for the call-to-action, e.g. \"Try Acme Scheduler free\". Pass null to clear."
},
"factCheck": {
"type": "object",
"properties": {
"groundingFallback": {
"type": "boolean",
"description": "Default true. When no first-party or site material matches a topic, write it from verified external sources. false = hold such topics instead (reason shown on the site)."
},
"minPrimarySources": {
"type": "integer",
"maximum": 10,
"minimum": 0,
"description": "Default 2. How many distinct external sites the reviewer must have fetched; lower it (e.g. 1) for niches with few authoritative sources."
},
"requireFirstParty": {
"type": "boolean",
"description": "Default true. false = do not hold topics lacking first-party material."
}
},
"description": "Fact-check and originality gate. requireFirstParty (default true) grounds each topic in first-party material from the owner, same-origin reference pages, or passages harvested automatically from the pages of the site itself; when none match, groundingFallback (default true) writes the topic from verified external sources (at least minPrimarySources of them) instead of skipping it. requireFirstParty=false turns the gate off (the other editorial checks still apply). Any hold is reported as lastGenerationHold on autoseo_list_sites."
},
"authorMode": {
"enum": [
"organization",
"person"
],
"type": "string",
"description": "Byline mode. \"organization\" (default) credits the organization honestly. \"person\" credits a named, attested author added in the dashboard; with no attested author the article is held, never published under an invented name."
},
"imageStyle": {
"type": "string",
"description": "Style guidance for generated post images, e.g. \"minimal flat illustration\""
},
"authorPages": {
"type": "boolean",
"description": "Set true once the integration kit author page is installed on the owner's own site, so articles link to it. Hosted blogs do not need this."
},
"imagePolicy": {
"enum": [
"owner_only",
"owner_or_stock",
"owner_stock_generated"
],
"type": "string",
"description": "Which images may be used. Default \"owner_or_stock\": the owner's own library, then licensed stock with credit. \"owner_stock_generated\" additionally allows an AI illustration, always labelled as AI-generated."
},
"pausedUntil": {
"type": [
"string",
"null"
],
"description": "Do not generate or publish AutoSEO articles for this site before this ISO date/datetime (e.g. \"2027-01-05\"). Pass null to clear. While paused, generate calls return 409 site_paused."
},
"productName": {
"type": "string",
"description": "Product-mentions layer: the customer's own product name, e.g. \"Acme Scheduler\". Pass null to clear."
},
"approvalMode": {
"enum": [
"draft",
"auto"
],
"type": "string",
"description": "\"draft\" (default) queues articles for approval; \"auto\" publishes articles that pass the automated checks with no approval (publish-capable token required)"
},
"publicBaseUrl": {
"type": "string",
"description": "git_pr connector ONLY: the real public base URL the site is deployed to (e.g. \"https://blog.acme.com\"), used to build the social-repurpose link. Without it, git-connected sites skip the blog->social repurpose (their publish result is a GitHub PR URL, never a live article)."
},
"questionMining": {
"type": "object",
"properties": {
"country": {
"type": "string",
"description": "Market for search data as an ISO country code such as us or es. Defaults to us."
},
"enabled": {
"type": "boolean",
"description": "Default false. Set true only when the user wants weekly question mining."
},
"language": {
"type": "string",
"description": "Search language code (default en)."
},
"autoAppend": {
"type": "integer",
"maximum": 5,
"minimum": 1,
"description": "How many new questions are appended to the topic list per weekly run. Default 3."
}
},
"description": "Opt-in question mining for this site. When enabled, at most once a week the engine pulls the real questions people ask (Google People Also Ask, related searches, public forum threads, and this site's own Search Console queries), checks measured volume, drops anything the site already covers, and appends the top autoAppend new questions to the topic list. Articles still follow the site approval mode; paused sites are skipped. OFF BY DEFAULT. Omit to leave the current setting unchanged."
},
"allowedLinkUrls": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 8,
"description": "Product-mentions layer: up to 8 https:// pages (features/pricing/docs) an in-article mention may link to instead of ctaUrl, whichever best matches that specific article. ctaUrl is implicitly allowed even when omitted here. Pass null to clear the whole list."
},
"backlinkTargets": {
"type": "array",
"items": {
"type": "object",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"description": "Backlink URL"
},
"relevance": {
"type": "string",
"description": "Why this target is relevant to the site"
}
}
},
"description": "URLs plus optional relevance notes to weave in as outbound backlinks from generated posts"
},
"editorialReview": {
"type": "object",
"required": [
"enabled"
],
"properties": {
"enabled": {
"type": "boolean",
"description": "Legacy compatibility field. Its value is ignored because editorial review is mandatory."
},
"instructions": {
"type": "string",
"maxLength": 6000,
"description": "Editorial expertise, factual constraints, and reader usefulness requirements."
},
"referenceUrls": {
"type": "array",
"items": {
"type": "string",
"format": "uri",
"pattern": "^https://"
},
"maxItems": 5,
"description": "HTTPS pages the reviewer should read to fact-check articles."
},
"verifiedFacts": {
"type": "array",
"items": {
"type": "object",
"required": [
"claim",
"sourceUrl"
],
"properties": {
"claim": {
"type": "string",
"maxLength": 2000,
"minLength": 1,
"description": "The factual statement you have verified, written as a complete sentence."
},
"notes": {
"type": "string",
"maxLength": 2000,
"description": "Caveats or context for the reviewer."
},
"sourceUrl": {
"type": "string",
"format": "uri",
"pattern": "^https://",
"maxLength": 2048,
"description": "Public page that supports the claim."
},
"validAsOf": {
"type": "string",
"maxLength": 100,
"description": "Date or version the claim was last checked against, e.g. 2026-11-01."
},
"jurisdiction": {
"type": "string",
"maxLength": 200,
"description": "Country or region the claim applies to, e.g. Spain, if it is law or rule dependent."
}
}
},
"maxItems": 20,
"description": "Claims you have already verified, each with its source, that articles may rely on."
},
"requireInternalLinks": {
"type": "boolean",
"description": "Require verified internal links; defaults to true when review is enabled."
}
},
"description": "Mandatory automated editorial check before publication. Failed or unavailable checks hold the article for review. The legacy enabled field is accepted for compatibility but ignored; review cannot be disabled or opted out of. This does not guarantee Google indexing or rankings."
},
"productOneLiner": {
"type": "string",
"description": "Product-mentions layer: a one-sentence description of what the product does, used both for the relevance check and as fallback copy in the generated mention. Pass null to clear."
},
"socialRepurpose": {
"type": "object",
"description": "Blog->social repurpose setting: whenever an article on this site publishes, auto-draft social announcements to the given platforms. Drafts wait for human approval unless autoPublish is true AND accountIds pins accounts for that platform, in which case they are scheduled ~10 minutes out. OFF BY DEFAULT. Omit entirely to leave the current setting unchanged. { enabled?: boolean (default false/OFF, opt-in), platforms?: (\"twitter\"|\"linkedin\"|\"linkedin_page\")[] (default twitter+linkedin once enabled), autoPublish?: boolean (default false), accountIds?: { [platform]: string[] } (pinned integration account ids) }"
}
}
},
"annotations": {
"title": "AutoSEO Configure",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
One switch for blog auto-publishing: set approvalMode "auto" (articles that pass the automated checks publish on cadence, no approval, no dashboard) or "draft" (every article waits for approval) on one or more sites (siteIds) or on EVERY connected site of the active org (all: true). Only ever touches the org this credential is bound to (check whoami first). "auto" needs a publish-capable token. Returns per site the stored approvalMode and effectiveApprovalMode (what the runner will really do; a search-recovery clamp can keep a site on draft, with the reason). CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs91/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
approvalMode
"auto" = publish without approval; "draft" = approval queue
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteIds
Site ids from autoseo_list_sites. Omit when all is true.
Clear enough to supply this argument.
all
true = every connected site in the active org. Omit when siteIds is given.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_set_approval_mode",
"description": "One switch for blog auto-publishing: set approvalMode \"auto\" (articles that pass the automated checks publish on cadence, no approval, no dashboard) or \"draft\" (every article waits for approval) on one or more sites (siteIds) or on EVERY connected site of the active org (all: true). Only ever touches the org this credential is bound to (check whoami first). \"auto\" needs a publish-capable token. Returns per site the stored approvalMode and effectiveApprovalMode (what the runner will really do; a search-recovery clamp can keep a site on draft, with the reason). CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"approvalMode"
],
"properties": {
"all": {
"type": "boolean",
"description": "true = every connected site in the active org. Omit when siteIds is given."
},
"siteIds": {
"type": "array",
"items": {
"type": "string"
},
"description": "Site ids from autoseo_list_sites. Omit when all is true."
},
"approvalMode": {
"enum": [
"draft",
"auto"
],
"type": "string",
"description": "\"auto\" = publish without approval; \"draft\" = approval queue"
}
}
},
"annotations": {
"title": "AutoSEO Set Approval Mode",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
Update the Build-in-Public automation (GitHub shipped-feature tweets) without the dashboard. mode "AUTO" posts every drafted event to X with no approval (needs a publish-capable token); "APPROVE" keeps each tweet in the approvals rail. Only the fields you pass change. Returns the saved configuration. Read the current state with build_in_public_status. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs84/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
mode
AUTO = post without approval; APPROVE = approval rail
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
enabled
Master switch: whether GitHub events produce tweet drafts at all.
Clear enough to supply this argument.
tone
Voice for generated tweets, e.g. "casual"
Clear enough to supply this argument.
dailyLimit
Most tweets drafted or posted per day in per-pr mode; 0 pauses posting.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
postMode
per-pr tweets about each merged change as it happens; digest batches changes into one summary tweet.
Clear enough to supply this argument.
digestFrequency
How often a digest tweet is produced when postMode is digest.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
digestDayOfWeek
Weekday the weekly digest goes out, counted from Sunday as 0.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
digestHourUtc
Hour of day, in UTC, the digest goes out.
Clear enough to supply this argument.
maxPostsPerWeek
Weekly cap on posts from this automation, across both modes.
Clear enough to supply this argument.
repoFilter
Comma-separated repos; null = all repos
Clear enough to supply this argument.
branchFilter
Branch to watch, e.g. "main"
Clear enough to supply this argument.
Annotations100/100
All four behavior hints are declared.
{
"title": "Build In Public Configure",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
Output schema0/100
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "build_in_public_configure",
"description": "Update the Build-in-Public automation (GitHub shipped-feature tweets) without the dashboard. mode \"AUTO\" posts every drafted event to X with no approval (needs a publish-capable token); \"APPROVE\" keeps each tweet in the approvals rail. Only the fields you pass change. Returns the saved configuration. Read the current state with build_in_public_status. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"properties": {
"mode": {
"enum": [
"AUTO",
"APPROVE"
],
"type": "string",
"description": "AUTO = post without approval; APPROVE = approval rail"
},
"tone": {
"type": "string",
"description": "Voice for generated tweets, e.g. \"casual\""
},
"enabled": {
"type": "boolean",
"description": "Master switch: whether GitHub events produce tweet drafts at all."
},
"postMode": {
"enum": [
"per-pr",
"digest"
],
"type": "string",
"description": "per-pr tweets about each merged change as it happens; digest batches changes into one summary tweet."
},
"dailyLimit": {
"type": "integer",
"maximum": 25,
"minimum": 0,
"description": "Most tweets drafted or posted per day in per-pr mode; 0 pauses posting."
},
"repoFilter": {
"type": [
"string",
"null"
],
"description": "Comma-separated repos; null = all repos"
},
"branchFilter": {
"type": "string",
"description": "Branch to watch, e.g. \"main\""
},
"digestHourUtc": {
"type": "integer",
"maximum": 23,
"minimum": 0,
"description": "Hour of day, in UTC, the digest goes out."
},
"digestDayOfWeek": {
"type": "integer",
"maximum": 6,
"minimum": 0,
"description": "Weekday the weekly digest goes out, counted from Sunday as 0."
},
"digestFrequency": {
"enum": [
"daily",
"weekly"
],
"type": "string",
"description": "How often a digest tweet is produced when postMode is digest."
},
"maxPostsPerWeek": {
"type": "integer",
"maximum": 100,
"minimum": 0,
"description": "Weekly cap on posts from this automation, across both modes."
}
}
},
"annotations": {
"title": "Build In Public Configure",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Act on one item from list_approvals, for ANY source, without the dashboard. Valid actions per source: posts: approve (schedules it) or reject. trends: approve (posts now), schedule (needs scheduledAt) or reject. build-in-public: approve (posts the tweet now) or reject. dms: approve (marks the DM approved), send (sends an approved DM) or reject (skips the lead). Returns the updated item. approve, schedule and send reach a live audience and need a publish-capable token. Only call approve/send when the human asked for it or has set this automation to run without approval. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs70/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
source
Which approval queue the item came from, as shown in its list_approvals entry.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
id
Item id from list_approvals
Clear enough to supply this argument.
action
What to do with the item. approve, schedule and send reach a live audience; which are valid depends on source.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
scheduledAt
Publish time as a future ISO 8601 datetime with timezone. Required for trends schedule; optional override for posts approve.
Clear enough to supply this argument.
reason
Optional rejection reason (posts)
Resolve the mismatch between this description and its schema constraints.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "resolve_approval",
"description": "Act on one item from list_approvals, for ANY source, without the dashboard. Valid actions per source: posts: approve (schedules it) or reject. trends: approve (posts now), schedule (needs scheduledAt) or reject. build-in-public: approve (posts the tweet now) or reject. dms: approve (marks the DM approved), send (sends an approved DM) or reject (skips the lead). Returns the updated item. approve, schedule and send reach a live audience and need a publish-capable token. Only call approve/send when the human asked for it or has set this automation to run without approval. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"source",
"id",
"action"
],
"properties": {
"id": {
"type": "string",
"description": "Item id from list_approvals"
},
"action": {
"enum": [
"approve",
"reject",
"schedule",
"send"
],
"type": "string",
"description": "What to do with the item. approve, schedule and send reach a live audience; which are valid depends on source."
},
"reason": {
"type": "string",
"description": "Optional rejection reason (posts)"
},
"source": {
"enum": [
"posts",
"trends",
"build-in-public",
"dms"
],
"type": "string",
"description": "Which approval queue the item came from, as shown in its list_approvals entry."
},
"scheduledAt": {
"type": "string",
"description": "Publish time as a future ISO 8601 datetime with timezone. Required for trends schedule; optional override for posts approve."
}
}
},
"annotations": {
"title": "Resolve Approval",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Switch an Instagram comment-to-DM funnel (from create_dm_funnel / list_dm_funnels) ON so matching comments get the private-reply DM automatically, or pass active:false to switch it off. Activation sends real DMs and needs a publish-capable token. On a post that has not published yet the funnel is armed (waitingForPublish:true) and goes live the moment the post publishes. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs84/100
Input documentation gives an agent enough context to populate the arguments.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "dm_funnel_activate",
"description": "Switch an Instagram comment-to-DM funnel (from create_dm_funnel / list_dm_funnels) ON so matching comments get the private-reply DM automatically, or pass active:false to switch it off. Activation sends real DMs and needs a publish-capable token. On a post that has not published yet the funnel is armed (waitingForPublish:true) and goes live the moment the post publishes. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"funnelId"
],
"properties": {
"active": {
"type": "boolean",
"description": "true (default) = activate; false = deactivate"
},
"funnelId": {
"type": "string",
"description": "Funnel id from list_dm_funnels"
}
}
},
"annotations": {
"title": "DM Funnel Activate",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description100/100
ONE-CLICK AutoSEO onboarding, "auto do the thing". For a connected site (siteId from autoseo_connect_site / autoseo_list_sites), this single call replaces the manual autoseo_propose_topics → autoseo_configure sequence: it (a) runs the topic-proposal engine to derive question-shaped topics/pillars for the site, (b) saves them into the site config with safe defaults for anything the user has not already set, weekly cadence, "illustration" image style, and approvalMode "draft" (the human approval gate; auto-configure NEVER opts a user into unattended publishing), and (c) if generateFirst is true, also queues the first article on the top proposed topic as a PENDING draft awaiting human approval (counts against the plan's monthly article quota; omit it to configure without generating). Existing deliberate settings are preserved: user-typed pillars stay first with proposed topics appended (deduped), and an existing cadence/imageStyle/approvalMode is never overwritten. Returns { config, proposedTopics (question + rationale, relay these to the user), firstPost? ({ status: "queued", draftId } on success, approve it with approve_draft after the human reviews it) }. Prefer this right after autoseo_connect_site when the user just wants it set up; use autoseo_propose_topics + autoseo_configure instead when they want to hand-pick topics. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
generateFirst
Also queue the first article (top proposed topic) as a pending draft behind the human approval gate. Uses one article of the monthly quota.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_auto_configure",
"description": "ONE-CLICK AutoSEO onboarding, \"auto do the thing\". For a connected site (siteId from autoseo_connect_site / autoseo_list_sites), this single call replaces the manual autoseo_propose_topics → autoseo_configure sequence: it (a) runs the topic-proposal engine to derive question-shaped topics/pillars for the site, (b) saves them into the site config with safe defaults for anything the user has not already set, weekly cadence, \"illustration\" image style, and approvalMode \"draft\" (the human approval gate; auto-configure NEVER opts a user into unattended publishing), and (c) if generateFirst is true, also queues the first article on the top proposed topic as a PENDING draft awaiting human approval (counts against the plan's monthly article quota; omit it to configure without generating). Existing deliberate settings are preserved: user-typed pillars stay first with proposed topics appended (deduped), and an existing cadence/imageStyle/approvalMode is never overwritten. Returns { config, proposedTopics (question + rationale, relay these to the user), firstPost? ({ status: \"queued\", draftId } on success, approve it with approve_draft after the human reviews it) }. Prefer this right after autoseo_connect_site when the user just wants it set up; use autoseo_propose_topics + autoseo_configure instead when they want to hand-pick topics. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId"
],
"properties": {
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"generateFirst": {
"type": "boolean",
"description": "Also queue the first article (top proposed topic) as a pending draft behind the human approval gate. Uses one article of the monthly quota."
}
}
},
"annotations": {
"title": "AutoSEO Auto Configure",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description92/100
Use the connected site name and configured pillars to brainstorm AI-estimated, question-shaped topics with rationale (search intent and AI-citation angle). This does NOT crawl the site or inspect existing pages. Questions already mined for the site (autoseo_mine_questions) come first, labelled with their source and measured volume; the rest are AI-estimated, not measured search data. Call after autoseo_connect_site and autoseo_verify_site, and before autoseo_configure, so the user approves a concrete topic list instead of a blind cadence. Relay the returned proposals to the user for approval, then pass the approved subset into autoseo_configure's topics. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs91/100
Input documentation gives an agent enough context to populate the arguments.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_propose_topics",
"description": "Use the connected site name and configured pillars to brainstorm AI-estimated, question-shaped topics with rationale (search intent and AI-citation angle). This does NOT crawl the site or inspect existing pages. Questions already mined for the site (autoseo_mine_questions) come first, labelled with their source and measured volume; the rest are AI-estimated, not measured search data. Call after autoseo_connect_site and autoseo_verify_site, and before autoseo_configure, so the user approves a concrete topic list instead of a blind cadence. Relay the returned proposals to the user for approval, then pass the approved subset into autoseo_configure's topics. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId"
],
"properties": {
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
}
}
},
"annotations": {
"title": "AutoSEO Propose Topics",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Mine the REAL questions people ask for a connected site and rank them. Sources: Google People Also Ask boxes, related searches, public forum threads (reddit, quora, stack exchange and community forums, with the thread url as evidence) and, when Search Console is connected, queries the site already appears for but ranks beyond position 8 on with no article. Each question is checked for measured monthly volume and difficulty, near-duplicates are merged, anything the site already covers is dropped, and the list is ranked by volume x (1 - difficulty) with a bonus for Search Console and forum demand. Results are saved as a keyword set (usable with keyword_add_to_plan). Pass addToPlan to append the top N new questions to the site topic list; articles still go through the normal approval gate. Paid provider calls are capped (a few seed queries per run, one volume lookup) and a site can be mined once an hour. Prefer this over autoseo_propose_topics when you want measured demand rather than an estimate. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Remove repeated schema validation from descriptions; focus on meaning and usage context.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
seeds
Optional starting queries to expand into real questions. Defaults to the site's own topics.
Clear enough to supply this argument.
country
Market for search data as an ISO country code such as us or es. Defaults to us.
Clear enough to supply this argument.
language
Search language as a code such as en or es. Defaults to en.
Clear enough to supply this argument.
addToPlan
How many of the top new questions to append to the site topic list. Omit or 0 to only review the ranked list.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_mine_questions",
"description": "Mine the REAL questions people ask for a connected site and rank them. Sources: Google People Also Ask boxes, related searches, public forum threads (reddit, quora, stack exchange and community forums, with the thread url as evidence) and, when Search Console is connected, queries the site already appears for but ranks beyond position 8 on with no article. Each question is checked for measured monthly volume and difficulty, near-duplicates are merged, anything the site already covers is dropped, and the list is ranked by volume x (1 - difficulty) with a bonus for Search Console and forum demand. Results are saved as a keyword set (usable with keyword_add_to_plan). Pass addToPlan to append the top N new questions to the site topic list; articles still go through the normal approval gate. Paid provider calls are capped (a few seed queries per run, one volume lookup) and a site can be mined once an hour. Prefer this over autoseo_propose_topics when you want measured demand rather than an estimate. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId"
],
"properties": {
"seeds": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 10,
"description": "Optional starting queries to expand into real questions. Defaults to the site's own topics."
},
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"country": {
"type": "string",
"description": "Market for search data as an ISO country code such as us or es. Defaults to us."
},
"language": {
"type": "string",
"description": "Search language as a code such as en or es. Defaults to en."
},
"addToPlan": {
"type": "integer",
"maximum": 10,
"minimum": 0,
"description": "How many of the top new questions to append to the site topic list. Omit or 0 to only review the ranked list."
}
}
},
"annotations": {
"title": "AutoSEO Mine Questions",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description97/100
List search phrases in a connected site's niche that are new or growing fast while the results are still open, so the site can publish before larger sites arrive. Each phrase has a 0-100 score from growth, measured volume, how weak the results are and fit with the site, plus its sources (Search Console, Google Trends, mined questions). Returns the stored weekly list. Pass refresh true to run it now, at most once a day per site. Rows are saved in a keyword set: pass keywordSetId and keywordId to keyword_add_to_plan to plan one, or call autoseo_generate_post with the phrase as topic. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs91/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
refresh
Run a fresh scan now (once a day per site). Default false returns the stored list.
Clear enough to supply this argument.
seeds
Optional seed phrases for a refresh. Default: the site's own topics.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_rising_phrases",
"description": "List search phrases in a connected site's niche that are new or growing fast while the results are still open, so the site can publish before larger sites arrive. Each phrase has a 0-100 score from growth, measured volume, how weak the results are and fit with the site, plus its sources (Search Console, Google Trends, mined questions). Returns the stored weekly list. Pass refresh true to run it now, at most once a day per site. Rows are saved in a keyword set: pass keywordSetId and keywordId to keyword_add_to_plan to plan one, or call autoseo_generate_post with the phrase as topic. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId"
],
"properties": {
"seeds": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 5,
"description": "Optional seed phrases for a refresh. Default: the site's own topics."
},
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"refresh": {
"type": "boolean",
"description": "Run a fresh scan now (once a day per site). Default false returns the stored list."
}
}
},
"annotations": {
"title": "AutoSEO Rising Phrases",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Generate one AI-drafted, answer-shaped blog post for a topic on a connected site and land it as a draft awaiting approval (or a future schedule), this never auto-publishes on its own. schedule is an optional future ISO datetime; omit it to leave the post as an unscheduled draft. Runs the platform's dedup/uniqueness/quality gates before returning and includes jsonldValid in the response, reporting whether the generated Article/FAQPage structured data parses cleanly, that's what makes a post eligible for AI-answer-engine citation. Source topic from autoseo_propose_topics rather than guessing one. Writing style for any post content you draft: write like the account's human owner, not like an AI. NEVER use em dashes, en dashes or double hyphens; use a period, comma or colon instead. Skip hashtag spam (0-1 hashtags), filler openers ("Exciting news!") and rocket/sparkle emoji. Before drafting, read the account's recent posts with list_posts and match their voice, casing and rhythm. For X specifically: keep a calm, factual, receipts-not-rhetoric tone; X ranking buries combative or bait-y posts even when engagement is high, and a mute or block outweighs hundreds of likes. Never write engagement-bait. Use 0-1 hashtags on X (more than 2 trips spam detection). Structure threads as one hook tweet, then 5-7 self-contained value tweets, then a single CTA. Prefer native vertical video (9:16, roughly 45-90 seconds) when media is available; it substantially outreaches text. Never draft identical or near-identical content for two different accounts. These heuristics were verified July 2026 and decay; treat as current-best, not gospel. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Input documentation gives an agent enough context to populate the arguments.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
topic
Target topic/keyword/question, ideally sourced from autoseo_propose_topics
Clear enough to supply this argument.
schedule
Optional future ISO 8601 datetime with timezone to publish the draft. Omit to leave it an unscheduled draft awaiting approval.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_generate_post",
"description": "Generate one AI-drafted, answer-shaped blog post for a topic on a connected site and land it as a draft awaiting approval (or a future schedule), this never auto-publishes on its own. schedule is an optional future ISO datetime; omit it to leave the post as an unscheduled draft. Runs the platform's dedup/uniqueness/quality gates before returning and includes jsonldValid in the response, reporting whether the generated Article/FAQPage structured data parses cleanly, that's what makes a post eligible for AI-answer-engine citation. Source topic from autoseo_propose_topics rather than guessing one. Writing style for any post content you draft: write like the account's human owner, not like an AI. NEVER use em dashes, en dashes or double hyphens; use a period, comma or colon instead. Skip hashtag spam (0-1 hashtags), filler openers (\"Exciting news!\") and rocket/sparkle emoji. Before drafting, read the account's recent posts with list_posts and match their voice, casing and rhythm. For X specifically: keep a calm, factual, receipts-not-rhetoric tone; X ranking buries combative or bait-y posts even when engagement is high, and a mute or block outweighs hundreds of likes. Never write engagement-bait. Use 0-1 hashtags on X (more than 2 trips spam detection). Structure threads as one hook tweet, then 5-7 self-contained value tweets, then a single CTA. Prefer native vertical video (9:16, roughly 45-90 seconds) when media is available; it substantially outreaches text. Never draft identical or near-identical content for two different accounts. These heuristics were verified July 2026 and decay; treat as current-best, not gospel. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"siteId",
"topic"
],
"properties": {
"topic": {
"type": "string",
"description": "Target topic/keyword/question, ideally sourced from autoseo_propose_topics"
},
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"schedule": {
"type": "string",
"description": "Optional future ISO 8601 datetime with timezone to publish the draft. Omit to leave it an unscheduled draft awaiting approval."
}
}
},
"annotations": {
"title": "AutoSEO Generate Post",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Import ONE finished, already-reviewed article and schedule it on a connected site at publishAt. Skips AI generation and the AI editorial review (the content is treated as pre-reviewed, stored with source "import"). The scheduler delivers it through the site's existing publish connection at publishAt like any generated article, and the weekly cadence cap does not hold it. Deduped per site by slug: re-sending a live slug returns status "duplicate" and changes nothing, so retries are safe. Counts nothing against the generation quota. Use list_articles with source "import" and status "scheduled" to inspect the queue. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs87/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
title
Article title shown as the H1 and in listings.
Clear enough to supply this argument.
slug
Lowercase kebab-case URL slug, unique per site (a-z, 0-9, -). Deduped: a slug already live on the site is skipped.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
markdown
The finished article body in Markdown. If faq is given and a question is not already in the body, an FAQ section is appended so the FAQ markup matches visible text.
Clear enough to supply this argument.
description
Meta description and excerpt for search results and listings.
Clear enough to supply this argument.
faq › items › question
The question as a reader would ask it.
Clear enough to supply this argument.
faq › items › answer
Plain-text answer shown under the question and in the FAQ markup.
Clear enough to supply this argument.
publishAt
ISO 8601 datetime WITH timezone offset or Z, e.g. 2026-11-03T09:00:00+01:00. A past time publishes on the next scheduler tick.
Clear enough to supply this argument.
author
Byline name shown as the article author.
Clear enough to supply this argument.
canonicalSource
Optional absolute URL of the article if it already exists elsewhere; becomes the canonical URL
Clear enough to supply this argument.
tags
Optional tags
Explain the meaning of this value, not just its name.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_import_article",
"description": "Import ONE finished, already-reviewed article and schedule it on a connected site at publishAt. Skips AI generation and the AI editorial review (the content is treated as pre-reviewed, stored with source \"import\"). The scheduler delivers it through the site's existing publish connection at publishAt like any generated article, and the weekly cadence cap does not hold it. Deduped per site by slug: re-sending a live slug returns status \"duplicate\" and changes nothing, so retries are safe. Counts nothing against the generation quota. Use list_articles with source \"import\" and status \"scheduled\" to inspect the queue. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"siteId",
"title",
"slug",
"markdown",
"description",
"publishAt",
"author"
],
"properties": {
"faq": {
"type": "array",
"items": {
"type": "object",
"required": [
"question",
"answer"
],
"properties": {
"answer": {
"type": "string",
"description": "Plain-text answer shown under the question and in the FAQ markup."
},
"question": {
"type": "string",
"description": "The question as a reader would ask it."
}
}
},
"maxItems": 30,
"description": "Optional FAQ pairs, rendered visibly and emitted as FAQPage JSON-LD"
},
"slug": {
"type": "string",
"description": "Lowercase kebab-case URL slug, unique per site (a-z, 0-9, -). Deduped: a slug already live on the site is skipped."
},
"tags": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 50,
"description": "Optional tags"
},
"title": {
"type": "string",
"description": "Article title shown as the H1 and in listings."
},
"author": {
"type": "string",
"description": "Byline name shown as the article author."
},
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"markdown": {
"type": "string",
"description": "The finished article body in Markdown. If faq is given and a question is not already in the body, an FAQ section is appended so the FAQ markup matches visible text."
},
"publishAt": {
"type": "string",
"description": "ISO 8601 datetime WITH timezone offset or Z, e.g. 2026-11-03T09:00:00+01:00. A past time publishes on the next scheduler tick."
},
"description": {
"type": "string",
"description": "Meta description and excerpt for search results and listings."
},
"canonicalSource": {
"type": "string",
"description": "Optional absolute URL of the article if it already exists elsewhere; becomes the canonical URL"
}
}
},
"annotations": {
"title": "AutoSEO Import Article",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Import up to 50 finished, already-reviewed articles for one site in a single call; same behavior as autoseo_import_article for each item (skips AI generation and AI review, scheduled at each publishAt, no cadence cap, deduped by slug). One bad item never blocks the others: the response has a per-item result (scheduled, duplicate or rejected with reasons) and a summary. Safe to re-run the whole batch. Keep the request under about 8 MB of Markdown; send large libraries in several batches. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.
The description explains the capability, when to use it, and what to expect.
Inputs93/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
siteId
Site id from autoseo_connect_site / autoseo_list_sites
Clear enough to supply this argument.
articles › items › title
Article title shown as the H1 and in listings.
Clear enough to supply this argument.
articles › items › slug
Lowercase kebab-case URL slug, unique per site (a-z, 0-9, -). Deduped: a slug already live on the site is skipped.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
articles › items › markdown
The finished article body in Markdown. If faq is given and a question is not already in the body, an FAQ section is appended so the FAQ markup matches visible text.
Clear enough to supply this argument.
articles › items › description
Meta description and excerpt for search results and listings.
Clear enough to supply this argument.
articles › items › faq › items › question
The question as a reader would ask it.
Clear enough to supply this argument.
articles › items › faq › items › answer
Plain-text answer shown under the question and in the FAQ markup.
Clear enough to supply this argument.
articles › items › publishAt
ISO 8601 datetime WITH timezone offset or Z, e.g. 2026-11-03T09:00:00+01:00. A past time publishes on the next scheduler tick.
Clear enough to supply this argument.
articles › items › author
Byline name shown as the article author.
Clear enough to supply this argument.
articles › items › canonicalSource
Optional absolute URL of the article if it already exists elsewhere; becomes the canonical URL
Clear enough to supply this argument.
articles › items › tags
Optional tags
Explain the meaning of this value, not just its name.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "autoseo_import_batch",
"description": "Import up to 50 finished, already-reviewed articles for one site in a single call; same behavior as autoseo_import_article for each item (skips AI generation and AI review, scheduled at each publishAt, no cadence cap, deduped by slug). One bad item never blocks the others: the response has a per-item result (scheduled, duplicate or rejected with reasons) and a summary. Safe to re-run the whole batch. Keep the request under about 8 MB of Markdown; send large libraries in several batches. CAPABILITY: `publish`: this can reach a LIVE AUDIENCE. A draft-scoped token gets a 403 here; a human must approve at timetopost.co instead.",
"inputSchema": {
"type": "object",
"required": [
"siteId",
"articles"
],
"properties": {
"siteId": {
"type": "string",
"description": "Site id from autoseo_connect_site / autoseo_list_sites"
},
"articles": {
"type": "array",
"items": {
"type": "object",
"required": [
"title",
"slug",
"markdown",
"description",
"publishAt",
"author"
],
"properties": {
"faq": {
"type": "array",
"items": {
"type": "object",
"required": [
"question",
"answer"
],
"properties": {
"answer": {
"type": "string",
"description": "Plain-text answer shown under the question and in the FAQ markup."
},
"question": {
"type": "string",
"description": "The question as a reader would ask it."
}
}
},
"maxItems": 30,
"description": "Optional FAQ pairs, rendered visibly and emitted as FAQPage JSON-LD"
},
"slug": {
"type": "string",
"description": "Lowercase kebab-case URL slug, unique per site (a-z, 0-9, -). Deduped: a slug already live on the site is skipped."
},
"tags": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 50,
"description": "Optional tags"
},
"title": {
"type": "string",
"description": "Article title shown as the H1 and in listings."
},
"author": {
"type": "string",
"description": "Byline name shown as the article author."
},
"markdown": {
"type": "string",
"description": "The finished article body in Markdown. If faq is given and a question is not already in the body, an FAQ section is appended so the FAQ markup matches visible text."
},
"publishAt": {
"type": "string",
"description": "ISO 8601 datetime WITH timezone offset or Z, e.g. 2026-11-03T09:00:00+01:00. A past time publishes on the next scheduler tick."
},
"description": {
"type": "string",
"description": "Meta description and excerpt for search results and listings."
},
"canonicalSource": {
"type": "string",
"description": "Optional absolute URL of the article if it already exists elsewhere; becomes the canonical URL"
}
}
},
"maxItems": 50,
"minItems": 1,
"description": "The finished articles to schedule, each with its own publishAt. Items are processed independently."
}
}
},
"annotations": {
"title": "AutoSEO Import Articles (batch)",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description92/100
Research and persist keyword opportunities for a seed keyword or supplied site/URL. When a live search data source is connected, covered rows return source="live" with measured metrics. Uncovered rows and unavailable requests use AI ideation with source="estimated". Exact searchVolume, difficulty and cpc remain null on estimated rows. Always relay each row source and never describe estimated bands as measured search data. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs81/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
seed
The starting keyword, domain or URL to find opportunities for.
Clear enough to supply this argument.
seedKind
Whether seed is a search phrase (keyword) or a domain, URL or brand label (site). Site mode never crawls the site.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
country
Market for search data as an ISO country code such as us or es. Defaults to us.
Clear enough to supply this argument.
siteId
Optional connected AutoSEO site id to associate with this keyword set.
Clarify units, where to obtain this value, or conventions not already expressed in the schema.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "keyword_research",
"description": "Research and persist keyword opportunities for a seed keyword or supplied site/URL. When a live search data source is connected, covered rows return source=\"live\" with measured metrics. Uncovered rows and unavailable requests use AI ideation with source=\"estimated\". Exact searchVolume, difficulty and cpc remain null on estimated rows. Always relay each row source and never describe estimated bands as measured search data. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"seed",
"seedKind"
],
"properties": {
"seed": {
"type": "string",
"description": "The starting keyword, domain or URL to find opportunities for."
},
"siteId": {
"type": "string",
"description": "Optional connected AutoSEO site id to associate with this keyword set."
},
"country": {
"type": "string",
"description": "Market for search data as an ISO country code such as us or es. Defaults to us."
},
"seedKind": {
"enum": [
"keyword",
"site"
],
"type": "string",
"description": "Whether seed is a search phrase (keyword) or a domain, URL or brand label (site). Site mode never crawls the site."
}
}
},
"annotations": {
"title": "Keyword Research",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description95/100
Add selected persisted keyword rows to an existing AutoSEO content plan. Appends the keyword text to the selected site's config.topics and marks those rows in_plan; it does not generate or publish an article. Later AutoSEO generation still lands behind the normal human approval gate. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Input documentation gives an agent enough context to populate the arguments.
keywordSetId
Keyword set id returned by keyword_research.
Clear enough to supply this argument.
keywordIds
One or more keyword row ids from that same set.
Clear enough to supply this argument.
siteId
Connected AutoSEO site whose content plan receives the keywords.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "keyword_add_to_plan",
"description": "Add selected persisted keyword rows to an existing AutoSEO content plan. Appends the keyword text to the selected site's config.topics and marks those rows in_plan; it does not generate or publish an article. Later AutoSEO generation still lands behind the normal human approval gate. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"keywordSetId",
"keywordIds",
"siteId"
],
"properties": {
"siteId": {
"type": "string",
"description": "Connected AutoSEO site whose content plan receives the keywords."
},
"keywordIds": {
"type": "array",
"items": {
"type": "string"
},
"description": "One or more keyword row ids from that same set."
},
"keywordSetId": {
"type": "string",
"description": "Keyword set id returned by keyword_research."
}
}
},
"annotations": {
"title": "Add Keyword To Plan",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Turn RAW MATERIAL into platform-native post drafts. Give it one or more URLs, pasted text, or both, and TimeToPost fetches + extracts them, brings out the claims (every claim must be backed by a verbatim quote from the source, unquotable claims are deterministically dropped, so it will not invent facts), plans genuinely DIFFERENT angles across platforms (an X thread and a LinkedIn post never carry the same claims), drafts each one in a real native format, and runs length/groundedness/originality gates. The drafts land PENDING in the human approval queue, this NEVER publishes and never schedules on its own; a human (or approve_draft, after they say yes) is what turns a draft into a post. Returns 202 with an ingestId: poll get_source_status until status is READY (then read the drafts), EMPTY (the page had too little text, ask the user to paste the article text and call again with `text`) or FAILED. Only platforms this org has CONNECTED are drafted for. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs79/100
Explain the meaning of this value, not just its name.
urls
Public http(s) article or page URLs to read and draft from. Private or internal addresses are refused.
Clear enough to supply this argument.
text
Pasted raw text to draft from. Use this when the URL is paywalled/JS-rendered (get_source_status returns EMPTY) or when there is no URL at all.
Clear enough to supply this argument.
title
Optional title for the pasted text.
Explain the meaning of this value, not just its name.
platforms
Platforms to draft for. Defaults to every connected platform among the allowed ones.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
brand
Optional brand/niche attribution tag for per-play rollups.
Explain the meaning of this value, not just its name.
scheduleHint
When to publish once a human approves: "optimal" picks a time from the org's own best-times model, otherwise a future ISO 8601 datetime with timezone.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "create_posts_from_source",
"description": "Turn RAW MATERIAL into platform-native post drafts. Give it one or more URLs, pasted text, or both, and TimeToPost fetches + extracts them, brings out the claims (every claim must be backed by a verbatim quote from the source, unquotable claims are deterministically dropped, so it will not invent facts), plans genuinely DIFFERENT angles across platforms (an X thread and a LinkedIn post never carry the same claims), drafts each one in a real native format, and runs length/groundedness/originality gates. The drafts land PENDING in the human approval queue, this NEVER publishes and never schedules on its own; a human (or approve_draft, after they say yes) is what turns a draft into a post. Returns 202 with an ingestId: poll get_source_status until status is READY (then read the drafts), EMPTY (the page had too little text, ask the user to paste the article text and call again with `text`) or FAILED. Only platforms this org has CONNECTED are drafted for. STYLE: follow the server writing-style instructions: write like the human owner, read list_posts first, no em or en dashes, 0-1 hashtags. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"properties": {
"text": {
"type": "string",
"description": "Pasted raw text to draft from. Use this when the URL is paywalled/JS-rendered (get_source_status returns EMPTY) or when there is no URL at all."
},
"urls": {
"type": "array",
"items": {
"type": "string"
},
"description": "Public http(s) article or page URLs to read and draft from. Private or internal addresses are refused."
},
"brand": {
"type": "string",
"description": "Optional brand/niche attribution tag for per-play rollups."
},
"title": {
"type": "string",
"description": "Optional title for the pasted text."
},
"platforms": {
"type": "array",
"items": {
"enum": [
"twitter",
"linkedin",
"threads",
"facebook"
],
"type": "string"
},
"description": "Platforms to draft for. Defaults to every connected platform among the allowed ones."
},
"scheduleHint": {
"type": "string",
"description": "When to publish once a human approves: \"optimal\" picks a time from the org's own best-times model, otherwise a future ISO 8601 datetime with timezone."
}
}
},
"annotations": {
"title": "Create Posts From Source",
"readOnlyHint": false,
"openWorldHint": true,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description98/100
Poll one source ingest by id (from create_posts_from_source). Returns its status (QUEUED → FETCHING → BRIEFING → PLANNING → DRAFTING → READY, or EMPTY/FAILED), the evidence-backed brief, the angle plan, the per-item extraction result (an item marked "thin" means that page had under 200 words of real text and was deliberately excluded rather than hallucinated from), and the PENDING drafts it produced. A draft whose payload carries needsFactCheck=true could not be fully verified against the source, surface that to the human before they approve it. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs75/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
ingestId
Ingest id returned by create_posts_from_source
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "get_source_status",
"description": "Poll one source ingest by id (from create_posts_from_source). Returns its status (QUEUED → FETCHING → BRIEFING → PLANNING → DRAFTING → READY, or EMPTY/FAILED), the evidence-backed brief, the angle plan, the per-item extraction result (an item marked \"thin\" means that page had under 200 words of real text and was deliberately excluded rather than hallucinated from), and the PENDING drafts it produced. A draft whose payload carries needsFactCheck=true could not be fully verified against the source, surface that to the human before they approve it. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"ingestId"
],
"properties": {
"ingestId": {
"type": "string",
"description": "Ingest id returned by create_posts_from_source"
}
}
},
"annotations": {
"title": "Get Source Status",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
List this org's recent source ingests (newest first) with their status, planned platforms, draft counts and per-item extraction results. Read-only. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description94/100
Create an Instagram comment-to-DM funnel for one published or scheduled post (to set one up while creating a post, use schedule_post dmFunnel instead). The funnel matches normalized whole keywords/phrases and sends one official Private Reply containing exactly one rendered {{link}} within seconds of the comment; publicReplyTemplate is optional and OFF when omitted. The funnel is ON by default (pass active:false to save it paused); on a scheduled post it goes live the moment the post publishes. Switching on needs a publish-capable token: a draft-only token saves it paused and dm_funnel_activate (or the dashboard) turns it on. Selling: set promoCode (an ACTIVE Stripe promotion code, e.g. TIMETOPOST30). While click tracking is on, the short link redirects to the lead URL plus ?promo= so checkout pre-applies it (the legacy shortener also adds utm_source=instagram, utm_medium=dm, utm_campaign and utm_content). trackClicks defaults to true and sends a click-tracked go short link (https://go.<product>/<readable-slug>; funnel.trackedUrl and destinationUrl are that link) tagged by linkTags {slug, channel, campaign, accountHandle, postLabel}. list_dm_funnels then reports hits, sent and clicks. trackClicks false stores the choice on the funnel and sends leadUrl exactly as given: no redirect, no shortener and no added parameters. Stats then report tracking "off" and null clicks instead of zero. Keyword matching ignores case, width and apostrophes (IT'S TIME TO POST = its time to post), and variants that collapse are deduplicated. For two offers on one post (GUIDE -> guide, IT'S TIME TO POST -> discount) create two funnels with different keywords. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.
The description explains the capability, when to use it, and what to expect.
Inputs89/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
name
Human label for this funnel in dashboards and in the default utm_campaign.
Clear enough to supply this argument.
integrationId
Instagram integration id from list_integrations.
Clear enough to supply this argument.
scope
What the funnel listens to. Only a single post is supported today.
Clear enough to supply this argument.
postId
Published or scheduled Instagram post id from list_posts.
Clear enough to supply this argument.
keywords
Comment words or phrases that trigger the DM, e.g. ["CHECKLIST"]. Case, width and apostrophes are ignored and whole words only are matched.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
dmTemplate
Private reply text containing exactly one literal {{link}} placeholder.
Clear enough to supply this argument.
leadUrl
HTTPS destination substituted for {{link}}.
Clear enough to supply this argument.
publicReplyTemplate
Optional public comment reply. Omit to keep public replies off (recommended default).
Clear enough to supply this argument.
dailyCap
Most DMs this funnel sends per day, to protect the account.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
promoCode
Optional active Stripe promotion code appended as ?promo= so checkout pre-applies it.
Clear enough to supply this argument.
utmCampaign
Optional utm_campaign (slugified). Defaults to the funnel name.
Clear enough to supply this argument.
trackClicks
Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "create_dm_funnel",
"description": "Create an Instagram comment-to-DM funnel for one published or scheduled post (to set one up while creating a post, use schedule_post dmFunnel instead). The funnel matches normalized whole keywords/phrases and sends one official Private Reply containing exactly one rendered {{link}} within seconds of the comment; publicReplyTemplate is optional and OFF when omitted. The funnel is ON by default (pass active:false to save it paused); on a scheduled post it goes live the moment the post publishes. Switching on needs a publish-capable token: a draft-only token saves it paused and dm_funnel_activate (or the dashboard) turns it on. Selling: set promoCode (an ACTIVE Stripe promotion code, e.g. TIMETOPOST30). While click tracking is on, the short link redirects to the lead URL plus ?promo= so checkout pre-applies it (the legacy shortener also adds utm_source=instagram, utm_medium=dm, utm_campaign and utm_content). trackClicks defaults to true and sends a click-tracked go short link (https://go.<product>/<readable-slug>; funnel.trackedUrl and destinationUrl are that link) tagged by linkTags {slug, channel, campaign, accountHandle, postLabel}. list_dm_funnels then reports hits, sent and clicks. trackClicks false stores the choice on the funnel and sends leadUrl exactly as given: no redirect, no shortener and no added parameters. Stats then report tracking \"off\" and null clicks instead of zero. Keyword matching ignores case, width and apostrophes (IT'S TIME TO POST = its time to post), and variants that collapse are deduplicated. For two offers on one post (GUIDE -> guide, IT'S TIME TO POST -> discount) create two funnels with different keywords. CAPABILITY: `draft`: creates content that cannot reach an audience on its own. Safe for a draft-only agent token.",
"inputSchema": {
"type": "object",
"required": [
"name",
"integrationId",
"scope",
"postId",
"keywords",
"dmTemplate",
"leadUrl"
],
"properties": {
"name": {
"type": "string",
"description": "Human label for this funnel in dashboards and in the default utm_campaign."
},
"scope": {
"enum": [
"post"
],
"type": "string",
"description": "What the funnel listens to. Only a single post is supported today."
},
"active": {
"type": "boolean",
"description": "false = save paused. Default true (on)."
},
"postId": {
"type": "string",
"description": "Published or scheduled Instagram post id from list_posts."
},
"leadUrl": {
"type": "string",
"description": "HTTPS destination substituted for {{link}}."
},
"dailyCap": {
"type": "number",
"maximum": 50,
"minimum": 1,
"description": "Most DMs this funnel sends per day, to protect the account."
},
"keywords": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 20,
"minItems": 1,
"description": "Comment words or phrases that trigger the DM, e.g. [\"CHECKLIST\"]. Case, width and apostrophes are ignored and whole words only are matched."
},
"linkTags": {
"type": "object",
"properties": {
"note": {
"type": "string",
"maxLength": 200,
"description": "Free-text note stored with the short link."
},
"slug": {
"type": "string",
"pattern": "^[A-Za-z0-9][A-Za-z0-9_-]{0,63}$",
"description": "Readable short-link slug. Default <product>-<handle>-<post>."
},
"channel": {
"type": "string",
"maxLength": 32,
"description": "Platform label, default instagram."
},
"campaign": {
"type": "string",
"maxLength": 60,
"description": "Product or campaign label."
},
"postLabel": {
"type": "string",
"maxLength": 80,
"description": "Post id or label."
},
"accountHandle": {
"type": "string",
"maxLength": 80,
"description": "Account handle the post runs on."
}
},
"description": "Farm tags for the tracked go short link (https://go.<product>/<slug>).",
"additionalProperties": false
},
"promoCode": {
"type": "string",
"description": "Optional active Stripe promotion code appended as ?promo= so checkout pre-applies it."
},
"dmTemplate": {
"type": "string",
"description": "Private reply text containing exactly one literal {{link}} placeholder."
},
"trackClicks": {
"type": "boolean",
"description": "Track link clicks (default true). False sends leadUrl exactly as given: no redirect, no shortener, no added parameters."
},
"utmCampaign": {
"type": "string",
"description": "Optional utm_campaign (slugified). Defaults to the funnel name."
},
"integrationId": {
"type": "string",
"description": "Instagram integration id from list_integrations."
},
"publicReplyTemplate": {
"type": "string",
"description": "Optional public comment reply. Omit to keep public replies off (recommended default)."
}
},
"additionalProperties": false
},
"annotations": {
"title": "Create DM Funnel",
"readOnlyHint": false,
"openWorldHint": false,
"idempotentHint": false,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
List this organization's post-scoped Instagram DM funnels, including funnelStatus (live | waiting_for_publish | paused), account and post snapshots, keywords, today's reserved send usage, poll errors, and reconnect state. Read-only; switch a funnel on or off with dm_funnel_activate. Each funnel includes stats {hits, sent, clicks, uniqueClicks, tracking}, promoCode, utmCampaign, trackClicks and destinationUrl (the tracked short link when trackClicks is true; the exact leadUrl when trackClicks is false, with stats.tracking "off" and clicks null). CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs90/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
integrationId
Only list funnels of this Instagram integration, id from list_integrations.
Clear enough to supply this argument.
scope
Only list funnels with this scope. Only single-post funnels exist today.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
active
true lists only switched-on funnels; false only paused ones. Omit for both.
Clear enough to supply this argument.
cursor
Pagination cursor from the previous page nextCursor. Omit for the first page.
Clear enough to supply this argument.
limit
How many funnels to return per page.
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_dm_funnels",
"description": "List this organization's post-scoped Instagram DM funnels, including funnelStatus (live | waiting_for_publish | paused), account and post snapshots, keywords, today's reserved send usage, poll errors, and reconnect state. Read-only; switch a funnel on or off with dm_funnel_activate. Each funnel includes stats {hits, sent, clicks, uniqueClicks, tracking}, promoCode, utmCampaign, trackClicks and destinationUrl (the tracked short link when trackClicks is true; the exact leadUrl when trackClicks is false, with stats.tracking \"off\" and clicks null). CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"properties": {
"limit": {
"type": "number",
"maximum": 100,
"minimum": 1,
"description": "How many funnels to return per page."
},
"scope": {
"enum": [
"post"
],
"type": "string",
"description": "Only list funnels with this scope. Only single-post funnels exist today."
},
"active": {
"type": "boolean",
"description": "true lists only switched-on funnels; false only paused ones. Omit for both."
},
"cursor": {
"type": "string",
"description": "Pagination cursor from the previous page nextCursor. Omit for the first page."
},
"integrationId": {
"type": "string",
"description": "Only list funnels of this Instagram integration, id from list_integrations."
}
}
},
"annotations": {
"title": "List DM Funnels",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
List captured comment evidence and delivery outcomes for one DM funnel: commenter, comment id/text, matched keyword, rendered reply snapshots, timestamps, detailed status, and public dmStatus lifecycle including window_expired. Read-only failure-queue visibility; a hit means a matching commenter, not a verified link conversion. CAPABILITY: `read`: side-effect-free.
The description explains the capability, when to use it, and what to expect.
Inputs94/100
Input documentation gives an agent enough context to populate the arguments.
funnelId
Funnel id from list_dm_funnels.
Clear enough to supply this argument.
status
Only return hits with this delivery status, as reported in the status field of earlier results (e.g. failed). Omit for all.
Clear enough to supply this argument.
cursor
Pagination cursor from the previous page nextCursor. Omit for the first page.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "list_funnel_hits",
"description": "List captured comment evidence and delivery outcomes for one DM funnel: commenter, comment id/text, matched keyword, rendered reply snapshots, timestamps, detailed status, and public dmStatus lifecycle including window_expired. Read-only failure-queue visibility; a hit means a matching commenter, not a verified link conversion. CAPABILITY: `read`: side-effect-free.",
"inputSchema": {
"type": "object",
"required": [
"funnelId"
],
"properties": {
"limit": {
"type": "number",
"maximum": 100,
"minimum": 1,
"description": "How many hits to return per page."
},
"cursor": {
"type": "string",
"description": "Pagination cursor from the previous page nextCursor. Omit for the first page."
},
"status": {
"type": "string",
"description": "Only return hits with this delivery status, as reported in the status field of earlier results (e.g. failed). Omit for all."
},
"funnelId": {
"type": "string",
"description": "Funnel id from list_dm_funnels."
}
}
},
"annotations": {
"title": "List Funnel Hits",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
Loading score analytics…
Put your MCP to the test
A few tasks worth trying with this server, chosen from its reviewed tools. These have not been run.
01List PostsList posts for the active org.
02Build Campaign URLsCAPABILITY: `read`.
03Get Engagement SummaryAggregated engagement metrics (likes, comments, shares, views) for the active org.
04Public Best Posting Times PreviewReturn the free, source-labeled public best-time curve for a platform.
05Calculate EngagementCAPABILITY: `read`.
Want to run these tasks against your MCP server with us?
TimeToPost setup, tools, quality score, publishing readiness, and alternatives.
How do I add TimeToPost to Claude, ChatGPT, or Cursor?
Add TimeToPost to Claude, ChatGPT, or Cursor by connecting its remote MCP URL in each client's MCP settings. In Claude, add a custom connector. In ChatGPT, add an MCP app. In Cursor, add an MCP server under Settings. The MCP URL is https://api.timetopost.co/mcp.
Is TimeToPost MCP ready for ChatGPT and Claude?
As of November 15, 2026, TimeToPost scores 100/100 on Manufact's MCP publishing checklist. The scan meets ChatGPT publishing checks and meets Claude publishing checks. Passing these checks does not guarantee marketplace approval.
How clear is TimeToPost's MCP tool documentation?
As of November 15, 2026, TimeToPost scores 85/100 in Manufact's review of tool titles, descriptions, input and output schemas, and behavior annotations. The review covers 77 of 77 discovered tools. The quality score measures documentation clarity; runtime behavior and marketplace readiness require separate checks.
What are the best alternatives to TimeToPost?
As of November 16, 2026, alternatives to TimeToPost in marketing-seo on the Manufact MCP leaderboard include Notra, BulkPublish, Fuxux, HYPD AI for App Store Optimization, and HYPD AI for Google Search Console.
How can TimeToPost improve its MCP documentation?
geo_visibility_check is the lowest-scoring reviewed tool at 78/100. No output schema is declared. Add one for structured results so clients can interpret returned fields. Open its review for feedback and a copyable fix prompt, then rerun the documentation review after updating your tool metadata.
Not your MCP?
Review any public or OAuth MCP server for protocol compatibility, tool documentation, and ChatGPT and Claude publishing readiness. No account required.