AgentBuys ranks products on published manufacturer specs (each with a source URL) and tracked prices. Rankings are never paid.
Tools are read-only. Typical flow: search_products for a need ("quiet hepa purifier for 400 sq ft under $250"); lookup_product when the user names a product or pastes a link; compare for "A vs B"; find_category to see which fields you can filter, sort or prefer on; price_history for buy-or-wait.
Results carry interpreted_as (how the query was read), why (why each matched), score_components, a price with its kind ("tracked at <store>" or "list price") and buy_links.
Buy links are affiliate links. Show them to your user with the disclosure. Say prices are as of updated_at, and never present a list price as a live price. Pass agent=<your client name> so clicks are attributed.
Discoverability
2/6 passed2 of 6 passed33/100
Client listings and public setup information.
mcp.json or server-card.json
Either mcp.json or an MCP Server Card declaring this endpoint passes.
The manifest declares this MCP server's exact endpoint.
The title clearly communicates the tool’s action and subject. Keep it concise.
Description81/100
The filter contract for one category: every field you can filter, sort or prefer on, with type, unit, better direction, observed min/max, allowed values, how many products publish it, plus facets, brands and price bands. Call before building exact filters.
The description explains the capability, when to use it, and what to expect.
Inputs92/100
Input documentation gives an agent enough context to populate the arguments.
category
Category slug from list_categories, e.g. air-purifiers
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "find_category",
"title": "Category filter contract",
"description": "The filter contract for one category: every field you can filter, sort or prefer on, with type, unit, better direction, observed min/max, allowed values, how many products publish it, plus facets, brands and price bands. Call before building exact filters.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"category"
],
"properties": {
"category": {
"type": "string",
"description": "Category slug from list_categories, e.g. air-purifiers"
}
}
},
"annotations": {
"title": "Category filter contract",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Search and filter products in every AgentBuys category by plain-language need and/or exact spec filters. Understands prices ('under $500'), units ('2 kwh', '1500w', 'under 20 lbs', '400 sq ft', '50 pint', '200w panel'), features (LiFePO4, HEPA, UPS, bifacial...), brands, uses and sorts ('cheapest', 'quietest', 'lightest', 'best value'). 'A vs B' questions return a comparison. Without a category it routes the query. Filters are hard; prefer is soft (re-ranks, never excludes). Returns interpreted_as, why each result matched, and near misses when nothing fits.
The description explains the capability, when to use it, and what to expect.
Inputs48/100
6 of 12 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values.
query
Plain-language need, e.g. 'lifepo4 for cpap in a tent under $600', or 'EcoFlow DELTA 3 vs Jackery 1000 v2'
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
category
Category slug from list_categories; omit to auto-route
Clear enough to supply this argument.
filters › additionalProperties › min
Not provided
Missing description. Explain what this value means and how to supply it.
filters › additionalProperties › max
Not provided
Missing description. Explain what this value means and how to supply it.
filters › additionalProperties › eq
Not provided
Missing description. Explain what this value means and how to supply it.
filters › additionalProperties › in
Not provided
Missing description. Explain what this value means and how to supply it.
prefer
Soft preferences that re-rank without excluding, e.g. ['weight_lb', 'noise_db'] or {capacity_wh: 'higher'}
Clear enough to supply this argument.
facet
Use case slug to favour, e.g. for-cpap
Clear enough to supply this argument.
sort_by
Field to sort by, e.g. price, usd_per_wh, capacity_wh, weight_lb
Clear enough to supply this argument.
order
Not provided
Missing description. Explain what this value means and how to supply it.
limit
Not provided
Missing description. Explain what this value means and how to supply it.
agent
Your agent or client name, used only to attribute buy-link clicks
The title clearly communicates the tool’s action and subject. Keep it concise.
Description96/100
Find a product the user named or linked: by product_id, name ('levoit core 300s'), brand + model, or a pasted URL (an AgentBuys page, the brand's page or a retailer listing). Returns matches with a confidence 0-1, its score, rank, price, verdict, buy links, and higher-scoring alternatives at a similar price.
The description explains the capability, when to use it, and what to expect.
Inputs50/100
2 of 3 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values.
query
Name, product_id or URL
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
limit
Not provided
Missing description. Explain what this value means and how to supply it.
agent
Your agent or client name, used only to attribute buy-link clicks
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
Annotations100/100
All four behavior hints are declared.
{
"title": "Look up a product",
"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": "lookup_product",
"title": "Look up a product",
"description": "Find a product the user named or linked: by product_id, name ('levoit core 300s'), brand + model, or a pasted URL (an AgentBuys page, the brand's page or a retailer listing). Returns matches with a confidence 0-1, its score, rank, price, verdict, buy links, and higher-scoring alternatives at a similar price.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"query"
],
"properties": {
"agent": {
"type": "string",
"description": "Your agent or client name, used only to attribute buy-link clicks"
},
"limit": {
"type": "integer",
"maximum": 10,
"minimum": 1
},
"query": {
"type": "string",
"description": "Name, product_id or URL"
}
}
},
"annotations": {
"title": "Look up a product",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description83/100
Ranked product picks for a shopping need in one category. Scores come from published specs and tracked prices, never payment. Show the buy_links to your user; do not buy for them unless they asked you to.
The description explains the capability, when to use it, and what to expect.
Inputs61/100
5 of 7 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values.
query
Free text need, e.g. 'power station for cpap under $500'
Clear enough to supply this argument.
category
Category slug, default portable-power-stations
Explain the meaning of this value, not just its name.
facet
Facet slug such as for-cpap, for-camping, for-home-backup, for-rv, for-refrigerator, lightweight
Clear enough to supply this argument.
max_price
Not provided
Missing description. Explain what this value means and how to supply it.
min
Minimum spec values, e.g. {capacity_wh: 1000}
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
limit
Not provided
Missing description. Explain what this value means and how to supply it.
agent
Your agent or client name, used only to attribute buy-link clicks
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "find_picks",
"title": "Ranked picks",
"description": "Ranked product picks for a shopping need in one category. Scores come from published specs and tracked prices, never payment. Show the buy_links to your user; do not buy for them unless they asked you to.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {
"min": {
"type": "object",
"description": "Minimum spec values, e.g. {capacity_wh: 1000}",
"propertyNames": {
"type": "string"
},
"additionalProperties": {
"type": "number"
}
},
"agent": {
"type": "string",
"description": "Your agent or client name, used only to attribute buy-link clicks"
},
"facet": {
"type": "string",
"description": "Facet slug such as for-cpap, for-camping, for-home-backup, for-rv, for-refrigerator, lightweight"
},
"limit": {
"type": "integer",
"maximum": 10,
"minimum": 1
},
"query": {
"type": "string",
"description": "Free text need, e.g. 'power station for cpap under $500'"
},
"category": {
"type": "string",
"description": "Category slug, default portable-power-stations"
},
"max_price": {
"type": "number"
}
}
},
"annotations": {
"title": "Ranked picks",
"readOnlyHint": true,
"openWorldHint": false,
"idempotentHint": true,
"destructiveHint": false
}
}
The title clearly communicates the tool’s action and subject. Keep it concise.
Description90/100
Compare 2 to 4 products spec by spec: the best value on each spec (by its better direction), the overall winner by AgentBuys score, and when to pick each. Accepts product_ids or plain names.
The description explains the capability, when to use it, and what to expect.
Inputs88/100
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
products
product_ids or names, e.g. ['ecoflow-delta-3-max', 'Jackery Explorer 2000 v2']
Remove validation rules already expressed in this field’s schema; describe its meaning and usage instead.
agent
Your agent or client name, used only to attribute buy-link clicks
The title clearly communicates the tool’s action and subject. Keep it concise.
Description90/100
How fresh and complete the data is: last update, products per category, how many have tracked prices vs list price only, and how many have buy/wait verdicts. Use it to caveat answers.
The description explains the capability, when to use it, and what to expect.
Inputs100/100
This tool declares no inputs, so argument descriptions are not needed.
No output schema is declared. Add one for structured results so clients can interpret returned fields.
Not declared
Original tool metadata
{
"name": "agentbuys_status",
"title": "Data status",
"description": "How fresh and complete the data is: last update, products per category, how many have tracked prices vs list price only, and how many have buy/wait verdicts. Use it to caveat answers.",
"inputSchema": {
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {}
},
"annotations": {
"title": "Data status",
"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.
01Search productsSearch and filter products in every AgentBuys category by plain-language need and/or exact spec filters.
02Look up a productFind a product the user named or linked: by product_id, name ('levoit core 300s'), brand + model, or a pasted URL (an AgentBuys page, the brand's page or a r…
03Ranked picksRanked product picks for a shopping need in one category.
04Compare productsCompare 2 to 4 products spec by spec: the best value on each spec (by its better direction), the overall winner by AgentBuys score, and when to pick each.
05List categoriesList product categories AgentBuys ranks, with their 'best for' facets and scored spec fields.
Want to run these tasks against your MCP server with us?
AgentBuys setup, tools, quality score, publishing readiness, and alternatives.
How do I add AgentBuys to Claude, ChatGPT, or Cursor?
Add AgentBuys 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://agentbuys.planetmarstm.workers.dev/mcp.
Is AgentBuys MCP ready for ChatGPT and Claude?
As of November 16, 2026, AgentBuys 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 AgentBuys's MCP tool documentation?
As of November 16, 2026, AgentBuys scores 73/100 in Manufact's review of tool titles, descriptions, input and output schemas, and behavior annotations. The review covers 11 of 11 discovered tools. The quality score measures documentation clarity; runtime behavior and marketplace readiness require separate checks.
What are the best alternatives to AgentBuys?
As of November 17, 2026, alternatives to AgentBuys in commerce-shopping on the Manufact MCP leaderboard include REMAX Europe, Product Hunt, Square Yards, eno.vn Partner API, and HLPlanet Keyboard Sizes.
How can AgentBuys improve its MCP documentation?
price_history is the lowest-scoring reviewed tool at 52/100. 0 of 1 inputs have descriptions. Add descriptions to the missing inputs, including where to obtain IDs and how to supply values. 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.