Check if a product feed is ready for AI shopping (ChatGPT, Gemini, AI Mode): score, gaps, fixes.
Scored across 3 tools
The three tools target clearly distinct purposes: scan_product_feed performs diagnosis, get_feed_fix_guide explains remediation, and list_ai_shopping_platforms provides background context. Their descriptions explicitly delimit scope (e.g. 'Overview only, not a list of live integrations'), leaving little room for misselection.
All three names follow a clean verb_noun pattern (get_, list_, scan_) with snake_case throughout. The naming is predictable and each name telegraphs its action and resource.
Three tools is on the lean side, but the server has a deliberately narrow purpose (auditing a product feed for AI shopping readiness) and each tool earns its place in the scan → understand → fix flow. Slightly under what a full lifecycle might want, but not thin enough to be a problem.
The surface covers the core loop of diagnosing a feed, explaining gaps, and pointing to fixes, with a platform overview for context. Minor gaps remain: no way to re-scan and compare after fixes or retrieve per-product detail beyond the stored summary.
Explain what a gap from scan_product_feed means, why AI shopping assistants need that data, and how to fix it with evidence the merchant already owns. Call with no argument to get the full checklist.
| Name | Required | Description | Default |
|---|---|---|---|
| issue_key | No | Issue key from scan_product_feed, e.g. no_specs, thin_desc, llm_unanswerable. Omit for all. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds that output is explanatory and backed by 'evidence the merchant already owns', which is useful color, but says nothing about response size, pagination, or whether the full checklist is large.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Roughly two sentences, front-loaded with the core purpose, and the calling convention is placed last as an actionable trailer. No filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With a single optional parameter, annotations covering the safety profile, and no output schema, the description supplies enough to call it correctly and sets expectations about the nature of the return. It could add a word on output volume or format, which keeps it from a 5.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and the schema already documents issue_key with concrete examples (no_specs, thin_desc, llm_unanswerable). The description only restates the omit-for-all behavior, so the baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb (explain) and resource (a gap from scan_product_feed and its fix), and explicitly ties the tool to the sibling that produces the input. An agent can tell it apart from scan_product_feed and list_ai_shopping_platforms without opening a schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It states the calling convention clearly ('Call with no argument to get the full checklist') and implies the input comes from scan_product_feed, giving a usable when-to-use context. It stops short of an explicit when-not-to-use or a named alternative, so it lands just under full marks.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List the main AI shopping surfaces and how product data reaches each one (e.g. Google Merchant Center for Gemini / AI Mode / Shopping; ChatGPT merchant program). Overview only, not a list of live integrations.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false; the description reinforces this by framing the tool as a static overview rather than querying live integrations, which is useful confirming context. It stops short of describing output shape or whether content is static/curated.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tightly packed sentences: the action and content come first, the scope-limiting caveat second. No filler, no repetition of the title, and every clause carries information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no parameters, no output schema, and annotations covering the safety profile, the description needs only to convey what a caller gets back, which it does ('the main AI shopping surfaces and how product data reaches each one'). Slightly better would be a note on static vs. refreshed content, but nothing essential is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Zero input parameters, so no parameter semantics are required and the baseline for a 0-param tool applies. The description correctly adds no fabricated input guidance.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (list) and resource (AI shopping surfaces) with concrete examples of what's covered (Google Merchant Center, ChatGPT merchant program). It scopes itself against a false expectation ('overview, not live integrations'), though it never names or contrasts with the feed-focused siblings get_feed_fix_guide/scan_product_feed.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The clause 'Overview only, not a list of live integrations' gives a clear when-not boundary that tells an agent this is reference/educational content, not a scan of the user's own data. It does not, however, explicitly say when to prefer it over the sibling feed tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch a public product feed once and score how well AI shopping assistants can understand it (0-100). Returns the biggest gaps (share of products affected), sample shopper questions AI could not answer from the feed, the thinnest products, and a link to fix the gaps. Accepts Google Merchant / Facebook catalog XML, CSV, TSV, or a Shopify store URL. Takes 20-60 seconds. Only summary results are stored.
| Name | Required | Description | Default |
|---|---|---|---|
| locale | No | Language for shopper questions and labels (default en). | |
| feed_url | Yes | Public URL of the product feed or Shopify store, e.g. https://example.com/feed.xml or https://shop.example.com |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=true, so the safety profile is covered. The description adds real operational context beyond that: the 20-60 second runtime and the retention note that only summary results are stored, plus the fact that it fetches externally. No auth or rate-limit details, but the latency and storage disclosures are genuinely useful.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Five short sentences, each carrying distinct information: action, return contents, accepted formats, latency, storage. Front-loaded with the action and no filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description carries the burden of describing returns and does so concretely (score, gaps with affected share, unanswerable questions, thinnest products, fix link). Combined with accepted formats and runtime, an agent has everything needed to call it correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3, but the description adds format semantics the schema omits: accepted inputs are Google Merchant / Facebook catalog XML, CSV, TSV, or a Shopify store URL. That clarifies what feed_url actually accepts, useful for a free-form URL parameter.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a concrete verb+resource (fetch a product feed and score AI shopping readiness 0-100) and immediately enumerates what the output contains (gaps, sample questions, thinnest products, fix link). An agent can distinguish this from get_feed_fix_guide and list_ai_shopping_platforms without opening a schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied by the described inputs and outputs, but there is no explicit when-to-use/when-not guidance and no named alternative. It references 'a link to fix the gaps' — which clearly relates to get_feed_fix_guide — yet never routes the agent to it, leaving the scan-then-fix handoff to inference.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_feed_fix_guidelist_ai_shopping_platformsscan_product_feedIs a website ready for AI shopping agents? Readiness score (0-100) + agent shopping simulation.
Preview-first feed optimization for retailers and agencies: audit, fix and serve product feeds.
Score a Magento or Adobe Commerce store on eight AI agent-readiness checks.
Grow sales on Google and beyond. Fix your Shopify product feeds and optimize every field in chat.
Add one secure layer between your agents and this server.
