Website audits for AI agents: 69 checks with exact fixes, plus verify. 5 free a day, then x402.
Scored across 4 tools
The tools mostly have distinct roles: full audit, customer-perspective review, check catalog, and before/after verification. The only mild overlap is audit_site versus verify_fixes, since verify_fixes runs a fresh audit internally, but the comparison use case and required audit_id keep them separable.
Three tools follow a clear verb_noun snake_case pattern: audit_site, list_checks, verify_fixes. customer_reaction is a noun phrase rather than verb_noun, a minor deviation that does not obscure its purpose.
Four tools is well-scoped for a focused site-audit server. Each tool has a distinct job: run an audit, list checks, gather simulated customer feedback, and verify fixes.
The surface covers the main lifecycle: audit, understand checks, get qualitative feedback, and verify remediation. Minor gaps remain, such as retrieving an old audit by audit_id or listing prior audits, but core agent workflows are covered.
Audit one public web page with 69 checks across security, SEO, TLS, performance, transport, accessibility, links, mobile, DNS and agent readiness. Returns the score, grade and the problems found, most important first, each with an exact fix, plus an audit_id to pass to verify_fixes after changes. Passive: only fetches the page and the files it links to. 5 free audits a day; after that it explains the paid API (0.01 USDC per audit via x402).
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Full http or https address of the page to audit, for example https://www.example.co.uk/ |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the readOnly/openWorld annotations, the description discloses what the tool actually does on the network ('passive: only fetches the page and the files it links to') and the rate limit with cost fallback ('5 free audits a day; after that... 0.01 USDC per audit via x402'). Rate limits and billing behavior are exactly the context annotations cannot carry.
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?
Three dense sentences: what it does, what it returns, then cost/rate limits. The most decision-relevant facts (scope, checks, passive behavior) are front-loaded and nothing is redundant with the schema.
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 compensates by describing the return payload (score, grade, ordered problems with exact fixes, audit_id). Combined with the rate-limit and passivity notes, an agent has everything needed to call it correctly and to chain into verify_fixes.
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 single url parameter is fully documented with an example, so the schema does the heavy lifting. The description still adds a constraint the schema does not state — the page must be 'public' — which meaningfully narrows valid inputs.
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 (audit) and resource (one public web page) with concrete scope: 69 checks across ten named categories. It also implicitly contrasts with the sibling by saying the audit_id is for verify_fixes, so the agent can distinguish the two without opening either 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?
The 'pass to verify_fixes after changes' sentence gives clear sequential routing to a sibling, and 'one public web page' scopes the input. There is no explicit when-not guidance (e.g., internal/authenticated pages, batch URLs), so it stops short of a full when/when-not/alternatives statement.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Eight AI-simulated customer perspectives read one homepage and report evidence-backed problems that would stop people buying: unclear prices, missing contact details, unexplained policies, weak proof. Each finding has a category, severity, the personas affected, evidence (a word-for-word quote checked against the page, or what is missing), why it matters and a suggested fix. No score. Pass previous_reaction_id to verify fixes: an unchanged page returns the earlier findings without using the free check. 1 free check a day; after that the paid API costs $0.05 via x402. Simulated perspectives, not real customers.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Full http or https address of the homepage, for example https://www.example.co.uk/ | |
| previous_reaction_id | No | Optional: the reaction_id from an earlier check of the same site, to verify fixes. Kept for 30 days. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the annotations it discloses free-tier quota (1/day), the $0.05 x402 fallback cost, that an unchanged page short-circuits to earlier findings, that reaction_id is retained 30 days, and that there is no score. The findings shape (category, severity, personas, evidence, why it matters, fix) tells the agent what will come back from a check-heavy operation.
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?
Three dense sentences, front-loaded with the core purpose before the output shape, quota, and simulated-perspective caveat. The enumeration of finding fields is long but earns its place because no output schema exists to carry it.
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 supplies the return shape itself, plus cost, quota, verification semantics, and the caveat that this is simulation. Nothing an agent needs to decide whether to call it or how to interpret results 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?
Schema coverage is 100%, so a 3 is the baseline, but the description earns extra credit by explaining what previous_reaction_id is for behaviorally (verify fixes; unchanged page returns earlier findings without consuming the free check) and by constraining url to a homepage. The schema covers syntax; the description covers intent.
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?
It states a specific verb and resource with unusual precision: eight simulated customer perspectives read one homepage and report evidence-backed buying blockers. It is clearly distinct from list_checks and audit_site, but it overlaps awkwardly with the verify_fixes sibling by claiming fix verification through previous_reaction_id without naming that alternative.
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 gives clear context for when to call it (one homepage, problems that would stop purchases) and describes the fix-verification path in concrete terms. It also caveats that the perspectives are simulated, not real customers. It never explicitly excludes cases or names a sibling tool for an adjacent job.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List every check AgentSiteAudit runs: its ID, category, severity and what it looks for, plus the category weights. Free and unlimited.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, openWorldHint=false, and destructiveHint=false, so the description doesn't need to repeat safety. It adds 'Free and unlimited', which communicates cost/rate behavior not captured by annotations, and describes the return contents since there is no output schema.
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?
A single tight sentence that front-loads the resource and enumerates return fields without waste. Every clause earns its place.
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 and no output schema, the description carries the full burden of explaining what comes back, and it does so by naming the returned fields and weights. Nothing an agent needs in order to call this correctly 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?
The tool takes zero parameters, so the baseline is 4. There is nothing parameter-related to explain, and the schema is complete for an empty object.
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 and resource ('List every check AgentSiteAudit runs') and enumerates the fields returned (ID, category, severity, criteria, category weights). This clearly distinguishes the metadata catalog from siblings like audit_site and verify_fixes, which act on sites rather than enumerate checks.
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 description implies usage as a discovery step before auditing (via 'what it looks for' and 'category weights'), but never explicitly says when to use this versus audit_site or verify_fixes. 'Free and unlimited' hints it is safe to call liberally, but no exclusions or alternatives are named.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Run a fresh audit of the same site and compare it with an earlier audit: lists what was fixed, improved, still failing, worse or new, with the score before and after. Use after making changes suggested by audit_site. Counts as one of the 5 free daily audits.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | The page to re-audit, on the same site as the earlier audit | |
| previous_audit_id | Yes | The audit_id returned by the earlier audit, like aud_01M6527XCRCB308T391BZGFZ9T. Audits are kept for 30 days. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover safety (readOnlyHint, destructiveHint=false) and openness, so the description's main value-add is the disclosed cost: "Counts as one of the 5 free daily audits" — a quota constraint an agent cannot infer from any structured field. The 30-day retention note appears in the schema rather than the description, and nothing is said about failure modes for expired IDs. No contradiction with readOnlyHint, since auditing does not mutate the target site.
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 sentences, front-loaded with the action, then the result categories, then the usage trigger, then the quota caveat. Every clause conveys distinct operational information with no restatement of the name or title.
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 only two required, fully documented parameters, annotations declaring safety, and no output schema, the description carries precisely what is missing: what the comparison returns, when to reach for it, and that it consumes daily quota. Nothing an agent needs in order to call it correctly is absent.
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 both parameters carry their own examples and patterns, so the schema does the heavy lifting. The description only reinforces the "same site" constraint on url; it adds no format or edge-case guidance for previous_audit_id beyond what the schema states.
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?
Specific verb and resource: "Run a fresh audit of the same site and compare it with an earlier audit," followed by the exact output taxonomy (fixed, improved, still failing, worse, new) plus before/after score. That output description cleanly separates it from audit_site without needing the 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?
Explicit trigger: "Use after making changes suggested by audit_site," which also implies audit_site is the tool for a first-time audit. It adds a real precondition in the schema-adjacent constraint of the page being "on the same site as the earlier audit," but never states a when-not case (e.g. what to do if no prior audit exists).
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.
customer_reactionaudit_sitelist_checksverify_fixesWebsite intelligence audits for AI agents: free preview plus x402-paid Agent Readiness scoring
The website checker for AI agents. One call: verdict, score out of 100, fix list.
Website audits for AI agents: GDPR consent, speed, WCAG, SPF/DMARC and AI-search readiness.
Free preflight and exact-price discovery for paid website and AI-agent audits.
Add one secure layer between your agents and this server.