Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=false, idempotentHint=false, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds genuine context beyond that: the 30-day full-widget trial, the included first scan, the returned site id and plan tier, and the authentication path (agent can self-register via 'Sign in with AgentID'). It does not explicitly state non-idempotency or duplicate-domain behavior, so it is not a 5.

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?

Front-loaded with the core action in the first sentence, followed by side effects, return values, and next steps in a compact block. Every sentence carries information, though the trailing instructions are slightly dense and could be split more cleanly.

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?

For a one-parameter creation tool with annotations covering the safety profile, the description supplies the missing pieces: prerequisites, side effects, and return contents (no output schema exists). Error conditions and rate limits are not covered, which keeps it below 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?

There is a single parameter with 100% schema description coverage, including an example URL and the 'one domain per site' constraint. The description adds no parameter-level detail beyond the schema, so the baseline 3 applies.

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 ('Add a website to the signed-in WebAbility account'), which cleanly separates it from siblings like list_sites, scan_page, or get_install_snippet. An agent can identify the operation without opening 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?

Provides clear workflow context by naming successors ('put the get_install_snippet tag on the site', 'use create_upgrade_link'), which effectively tells the agent where this tool sits in a sequence. It states the prerequisite (a free WebAbility account) but never says when NOT to use it, so it stops short of explicit exclusions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only, idempotent, non-destructive behavior, and the description adds substantial beyond that: error semantics (a non-matching selector or an HTTP-error page is an error, never 'no violations'), the per-rule node cap of 5 with nodesTotal + truncated flags, the nodeLimit ceiling of 50, and the escape hatch via scan_history. That is exactly the extra context the annotations do not provide.

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?

It is dense and front-loaded, leading with what is validated and which inputs select which mode, then layering error and truncation semantics. The parenthetical rule list and cap details make it a long single paragraph, but nearly every clause carries information an agent needs; only slight compression would help.

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?

There is no output schema, yet the description states what is returned (`violations`), how results can be truncated, and how to retrieve the full set via scan_history. Combined with the error-vs-empty distinction, an agent has everything needed to call and interpret this tool 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 real meaning: the html/url mutual exclusivity with its 'pass this or `html`' framing, the note that ARIA references outside the `selector` still resolve, and the default/max bounds on nodeLimit tied to truncation behavior. It does not add syntax examples or edge-case formats for the string 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?

The description names a specific verb (validate) plus precise resources (ARIA attribute + accessible name/role/value usage) and the two input modes (html snippet or live url). It also enumerates the underlying axe-core rule categories (cat.aria, cat.name-role-value) so an agent can distinguish it from scan_page, scan_html, and check_color_contrast 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 gives clear context: use with `html` for a snippet or `url` for a live page, optionally narrowed with `selector`, and it names scan_history(id) as the alternative when the truncated node set is insufficient. It does not explicitly say when to prefer this over sibling scanners like scan_html or scan_page, so it stops short of full alternative-routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial context beyond the annotations: the AA threshold values, AAA verdict only when level is 'AAA', error behavior for gradient/image backgrounds, brand-aligned replacement suggestions, and the scanner-based palette extraction. These details matter for correct use and interpretation.

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?

Dense but front-loaded: the core purpose and thresholds come first, then invocation modes, then edge cases and suggestions. Every clause adds useful information, though the single paragraph has many semicolon-separated ideas that could be slightly broken up.

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?

For an 8-parameter tool with no output schema and rich behavior, the description covers thresholds, input modes, precedence rules, error handling, and output shapes (ratio + pass/fail, AAA verdict, brand suggestions). It gives the agent enough to call the tool correctly and anticipate results.

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. The description goes further by explaining precedence between explicit colors and live-page extraction, the compositing of background from nearest painted ancestors, and that brandColors skips URL extraction, adding meaningful behavioral semantics to the parameters.

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: 'Check text contrast against the WCAG AA threshold (4.5:1, or 3:1 for large text)'. It also distinguishes itself from siblings by referencing the selector from a scan_page issue and by scoping the tool to contrast checking rather than page-wide auditing.

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?

Explicitly describes the two main invocation modes: pass a foreground/background pair, or pass url + selector to read colors from the live page. It also clarifies precedence ('explicit colors win over the page') and the fallback behavior when neither url nor brandColors is provided. No explicit when-not-to-use guidance or named alternative tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations mark this non-read-only and open-world, and the description adds real context beyond that: the link goes to a human, payment auto-activates the plan, card details never touch the agent, and an account/site prerequisite exists. It does not clarify whether each call mints a new link (non-idempotent) or how links expire, leaving a small gap.

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?

Front-loads the core action in the first sentence, then adds prerequisites and follow-up steps. Parentheticals ('the widget subscription', 'your owner') add clarity rather than waste, though the description could be tightened slightly.

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 usefully explains what is produced (a Stripe Checkout link), what happens after payment, and how to verify with list_sites. It omits what an agent should do if the user declines or the link expires, a minor gap for a two-parameter tool.

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 description coverage is 100%, so the schema already documents 'domain' and the 'billing_interval' enum with its monthly default. The description reinforces that the link targets one specific site but adds no format or edge-case detail beyond the schema. Baseline 3 applies.

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: gets a Stripe Checkout link that places one site on the WebAbility Pro plan. It scopes the action ('one site') and names the sibling tools involved (add_site, list_sites), so an agent can place it in the workflow without opening other schemas.

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?

Gives clear context: hand the link to the human who pays, the site must already exist on the account (add_site first), and a free WebAbility account is required. It also points to list_sites to confirm activation. There is no explicit 'do not use this when...' exclusion, so it stays just short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent, non-destructive and open-world behavior, so the safety profile is covered. The description adds real behavioral context beyond that: the 'unknown' fallback when no signal exists, that evidence is listed alongside the verdict, and that detection is shared/identical with scan_page's fields. It does not discuss failure modes for unreachable URLs, which keeps it short of a 5.

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?

One dense but well-structured sentence front-loads the verb and the two outputs, then chips in the evidence/interop note. Every clause (field names, example values, 'unknown' sentinel, scan_page equivalence) 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 only one parameter and no output schema, the description adequately substitutes by naming the two returned fields with representative values and the sentinel for no-signal cases, plus the downstream consumer. Nothing critical is missing for a low-complexity detection tool, though error/empty-result behavior is unaddressed.

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?

There is a single required `url` parameter already documented at 100% schema coverage ('URL to inspect'), so the schema carries the parameter burden. The description adds no format, protocol, or normalization details for the URL, which is the expected baseline when the schema does the heavy lifting.

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 ('Detect a page's stack') and enumerates the two returned fields with concrete example values, so the agent knows exactly what it obtains without opening a schema. It also distinguishes itself from scan_page by noting that sibling reports the same two fields from the same detection.

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?

Gives a clear use case: obtain the `framework` value to pass into generate_ai_fix, and clarifies overlap with scan_page rather than duplicating it. It stops short of stating explicit when-not conditions (e.g. use scan_html for raw HTML, or that scan_page is preferable when a full scan is already needed), so it is clear context but not full routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuinely non-obvious behavior beyond that: findings are matched by issue id (rule + element), and a changed class/id on a fixed element will appear as BOTH fixed and new — a false-regression trap an agent could not infer from the 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?

Front-loads the output shape (what changed) before the plumbing, and every clause carries information. It is dense — the matching-semantics sentence and the typical-loop sentence are both long — but no sentence is 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?

No output schema exists, so the description carries the return-value burden and does so by naming and defining fixed[]/new[]/remaining[]. Combined with the baseline/current input pairing and the matching caveat, an agent has enough to call it correctly without follow-up.

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 real meaning: it pairs the optional inputs (baseline side = baselineId or baselineUrl; current side = url or currentId) and notes baselineId is local-installs-only. That encoding of mutual alternatives goes beyond the flat per-parameter schema descriptions.

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 ('Compare two scans of the same page') and immediately enumerates the three output buckets (fixed/new/remaining). It explicitly distinguishes itself from the sibling verify_fix by scope ('Page-level complement to verify_fix (one element)'), so an agent can route between them 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?

Gives an explicit when-to-use recipe ('Typical loop: scan_page → edit → diff_scan(...) → confirm new[] is empty') plus the conditions selecting each input pair (baselineId vs baselineUrl, url vs currentId). It also warns to check new[] before declaring a regression, which is actionable guidance, not just context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so safety and idempotency are covered. The description adds file-type coverage (React/Vue/Svelte/HTML/PHP/Astro) but does not disclose return format, false-positive behavior, or performance characteristics. With annotations covering the safety profile, a 3 is appropriate.

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 sentences, each with a purpose: action, rationale, and file-type scope. No wasted words and the core use case is front-loaded. Could be slightly tighter but is well-structured.

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?

For a read-only lookup tool with full schema coverage and no output schema, the description provides enough context: what it does, why (edit the right file), and supported file types. Missing return format details, but those are minor for a read-only tool.

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 schema fully documents both parameters (selector and rootDir). The description reinforces 'given CSS selector' but adds no syntax or format details beyond the schema. Baseline 3 is correct when the schema does the heavy lifting.

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?

Clear verb ('Find') and resource ('source files') with a specific scope: files containing a given CSS selector. Distinguishes from siblings like scan_page or get_audit, but does not explicitly name an alternative tool for related tasks.

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 when to use it: mapping a DOM selector from a scan issue back to code. No explicit when-not-to-use or alternative tools are mentioned, so usage is only implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/openWorld/idempotent annotations: it discloses deterministic sequential execution (one page fully rendered before the next), single consolidated deduplicated report with the same fix payload as scan_page, and an enumerated per-URL outcome taxonomy including blocked bot-challenges. The guarantee that a page is "never silently dropped" is exactly the kind of behavior an agent needs and cannot get from annotations.

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?

Front-loaded with what it is, then how it walks, then what it returns, then the routing rule. Dense but every sentence carries information; the long parenthetical status list is slightly packed but each status is decision-relevant.

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?

For a 9-parameter tool with no output schema, the description still covers the return contract (one consolidated report, pages[] with explicit per-URL outcomes) and the invocation constraint on autoNavigate. Nothing essential for correct invocation 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 the baseline is 3; the description earns an extra point by adding execution semantics to autoNavigate (deterministic, sequential, headless server cannot discover the journey interactively) that the schema's terse 'URLs to walk after startUrl' does not convey.

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+resource (scan a multi-page user journey covering startUrl plus autoNavigate) and explicitly contrasts itself with the sibling scan_page. An agent can distinguish it from scan_page 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?

Explicitly names the alternative ("For a single page, use scan_page") and the selecting condition ("Better than per-page scans for journeys (login → checkout etc)"). No inference required to route between the two scan tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With annotations present, the description still adds real context beyond them: the count of alternatives returned per issue class, that brand palette is auto-extracted from the live URL when brandColors is omitted, and that output is ready-to-paste code per framework. It doesn't cover error modes or rate/scanning costs, keeping it from a 5.

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 sentences, front-loaded with the core action and then branching by issue type; each sentence carries information. Minor redundancy between 'brand-aligned' and the brand-palette explanation, but no wasted padding.

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?

No output schema exists, so the description carries the return-value burden and does so well (alternative counts, ready-to-paste code, per-framework targeting). With six schema-documented params and a live-scanning dependency, it is nearly complete, though it omits what happens on a framework/issue mismatch or scanning failure.

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 description coverage is 100%, so the baseline is 3. The description reinforces the brandColors/url interaction but adds little syntax or format detail beyond what the schema already documents for url, html, issue, context, framework and brandColors.

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 (generate) and resource (framework-aware fix alternatives) scoped to 'a specific accessibility issue,' and further narrows by issue type (color contrast vs label/ARIA). This is enough to separate it from siblings like verify_fix and check_color_contrast without opening any 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 description makes clear the context in which it applies (contrast vs label/ARIA issues, detected framework) and notes the brandColors/url fallback, so an agent knows when the tool is the right call. It stops short of explicitly naming alternatives or stating when not to use it, so it does not reach 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare the write/open-world/destructive profile, and the description adds meaningful context beyond them: local installs only, capped at 500 findings, free/no account, saved next to the project, and the grouping/scoring behavior. It does not explain the destructiveHint (e.g. whether an existing PDF is overwritten) or what a successful return looks like, so it stops short of a 5.

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?

Front-loaded with the core purpose, then input requirements, behavior, usage condition, limit, and alternative in a logical order. It is dense but nearly every sentence earns its place; only slight trimming would improve 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?

For a file-generating tool with no output schema, the description covers what is produced, where it is saved, and the 500-item cap, which is close to sufficient. It never explicitly states the return value (e.g. the saved file path), which is the one missing piece an agent would want.

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 genuine provenance: pass the issues[] array that a scan_page call returned, tying the parameter to a specific upstream tool. The 500-finding cap is also a relevant constraint on that parameter. widgetDetected is left to the schema.

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 (turn scan findings into) and resource (branded WebAbility accessibility-report PDF), plus the input it consumes and its output. It explicitly distinguishes itself from the sibling start_audit, so an agent can pick between them without opening schemas.

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?

Names the when-to-use condition (user wants a deliverable to attach to an email, ticket, or compliance thread) and the when-not/alternative (for the full audit deliverable, use start_audit instead). This is explicit routing rather than implied usage.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial context beyond the read-only annotations: ownership restriction, recommended polling cadence, the progress pipeline, and short-lived download URLs for JSON and Excel outputs. These are operationally important behaviors an agent needs before calling.

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?

Core purpose and polling guidance are clear and mostly front-loaded after the availability note. Minor redundancy appears with 'FREE' in the bracket and 'Free with a WebAbility account' repeated at the end.

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?

For an async audit-status tool with no output schema, the description covers return values, progress stages, completion artifacts, polling behavior, and access restrictions. The single input parameter is fully documented in the schema.

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 'id' parameter is already documented as the audit id returned by start_audit. The description reinforces that source but adds no syntax or format details beyond the schema, so 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?

States a specific verb ('Check') and resource ('an audit started with start_audit'), and describes exactly what it returns: overall status, per-step progress, and completion artifacts. This clearly distinguishes it from the sibling start_audit tool.

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?

Explicitly says to check an audit started with start_audit, gives polling guidance ('Poll every ~15s while status is pending/running'), and states an access exclusion ('Only the account that started an audit can read it'). It also explains the hosted MCP connection prerequisite.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover read-only, idempotent, non-destructive, closed-world behavior with no parameters. The description adds genuinely useful context beyond that: no account is needed, the widget resolves the site by domain, and the same tag works across all sites. It does not address caching or whether the tag changes over time, but the extra detail is solid.

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 tight sentences front-loaded with what the tag is, followed by placement and the domain/no-account facts. Every sentence earns its place with 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 input params and no output schema, the description fully covers what an agent needs: what the tool returns, where the snippet goes, and the key constraint that it is domain-based and account-free.

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 parameters, so nothing to document; baseline applies. The description compensates by explaining that the tag is domain-resolved rather than parameter-driven, which tells the agent why no input is required.

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 ('Get the one-line script tag') and names the exact artifact produced (the install tag for the WebAbility widget). It is distinguishable from siblings like add_site or list_sites without opening any 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?

Explains the context of use: place it in the <head> or before </body> of a site added with add_site. This ties invocation to a workflow step, though it gives no explicit when-not-to-use or alternate tool comparison.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so safety is covered. The description adds genuine context beyond that: rules span two engines, each carries a fixability tier and a fix op template, which tells the agent what differentiates the output. It omits ordering/pagination, keeping it at a 4 rather than 5.

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?

Front-loaded with the core purpose, then the value proposition and return fields in two tight sentences. The trailing return-field enumeration is slightly list-like, but it is useful given there is no output schema, so it earns its space.

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 usefully enumerates returned fields (ruleId, engine, tags/wcag, fixability, fix) and explains both engines and the filter dimensions. Minor gaps remain around result ordering or size limits, but nothing essential for correct invocation 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% and the schema descriptions are detailed (whole-word matching, tag formats, engine enum semantics), so the schema does the heavy lifting. The description only refers to 'optional filters' generically and adds no syntax or format meaning beyond the schema, matching the baseline 3.

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 opens with a specific verb and resource ('List accessibility rules') and scopes it precisely to two named engines with concrete counts (axe-core 104, WebAbility 90+). An agent can immediately distinguish this rule-catalog tool from scanning siblings like scan_page or check_aria.

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 a clear use context ('pick the rules worth auto-fixing before scanning'), which implies when to call it ahead of a scan. However, it never names an alternative tool or an explicit when-not-to-use case, so routing is inferred rather than stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent and non-destructive behavior, so the safety profile is covered. The description adds genuine context beyond them: it scopes results to the signed-in account and states the auth prerequisite (free account), which the annotations do not convey.

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 short sentences, front-loaded with what is returned, the use case, and the prerequisite. Every clause carries information and there is 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 parameters and no output schema, the description compensates by naming the fields returned (site id, plan tier, plan end date) and the auth requirement. An agent has everything needed to call it and interpret results.

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 there is nothing for the description to disambiguate; the baseline for a 0-param tool is 4. No misleading detail is added either.

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 (sites on the signed-in WebAbility account) and even enumerates the returned fields (site id, plan tier, plan end date). It is clearly distinguishable from the audit/scan siblings and from add_site.

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?

Gives an explicit use case ("check that a payment activated WebAbility Pro on a site") and a prerequisite ("Needs a free WebAbility account"), so the agent knows when to reach for it. It does not name a when-not condition or an alternative tool, so it stops short of 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/non-destructive/local-only, so the safety profile is covered. The description adds meaningful context beyond that: the on-disk storage path (~/.webability/scans) and the crucial caveat that the hosted server keeps no history, which materially affects expectations. It does not describe pagination or ordering beyond the default list behavior.

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?

Front-loaded with the core purpose, then progressively adds scope, storage caveat, and parameter behavior. Every sentence carries information an agent needs; no filler or repetition.

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 stating what the default list returns (when, target, result summary) and what `id` returns (full stored result). Combined with the storage-location caveat, 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 baseline is 3; the description earns above baseline by emphasizing that `id` returns the FULL stored result (not just a summary row) and that `filter` matches URL or tool substring. It adds real semantic shading beyond the schema, though it doesn't restate the `limit` default.

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 (browse past scans) and immediately scopes it by naming the exact sibling tools whose calls get logged (scan_page, flow_scan, scan_html, visual_audit, check_aria, verify_fix). An agent can distinguish this history/retrieval tool from the scan-execution siblings without opening any 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?

Gives clear mode-based guidance: no arguments lists recent scans, pass `id` for a full stored result, pass `filter` for substring matching. It does not name competing siblings (e.g. diff_scan, get_audit) or state when NOT to use it, so it falls short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the readOnly/idempotent annotations: it discloses the execution model (jsdom + WebAbility detectors + axe-core, no browser, no network), the fragment auto-wrapping behavior, and the critical limitation that visual-tier rules (contrast, target size, focus ring) are NOT evaluated with the dropped count surfaced as skippedVisual. This is exactly the kind of non-obvious behavioral context that prevents wrong invocation.

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?

Front-loads the core purpose before the execution details, and each sentence carries information. The heavy use of em-dash parentheticals makes it dense, but it is not padded with 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 correctly takes on the burden of explaining the return shape ('scan_page's shape (issues / summary) with fix.op + fixability'), plus the compact-vs-json output behavior and skippedVisual. An agent has enough to call it and interpret the result without opening anything else.

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 description coverage is already 100%, so the baseline is 3, but the description adds real meaning to engine (in-process vs browser tradeoff and rule coverage), format defaults, and the skippedVisual consequence of the engine choice. It reinforces and disambiguates the schema rather than merely repeating it.

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 ('Scan a raw HTML snippet or component markup') plus the key scoping constraint ('without serving it'), which distinguishes it from the served-page sibling scan_page. An agent can tell immediately what it does and roughly where it sits relative to scan_page.

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?

Clearly frames the default in-process path as fitting 'inside a tight edit loop' and states the condition for switching (pass engine:"browser" for visual-tier rules), which is explicit when-to-use guidance. It does not, however, explicitly name scan_page as the alternative for already-served URLs, so sibling routing is only implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With annotations already covering read-only/idempotent/open-world, the description still adds substantial behavioral detail beyond them: uncertain findings are 'judged or dropped, never listed for review', source pointers come only from React C18/Vue dev builds, and each issue carries a fixability tier explaining how far the agent may act on it. This is exactly the extra context the annotations cannot supply.

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?

Purpose is front-loaded in the first sentence, and the dense second sentence is justified because there is no output schema to carry the return-shape contract. It is on the long side and packs return semantics into a single block, but no sentence is 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?

For a 9-parameter tool with no output schema, the description covers the return shape (issues, summary, source, fix.op, fixability) unusually well and discloses how uncertain findings are handled. It omits operational facts such as whether protected/authenticated pages can be scanned, timeouts, or how results relate to scan_history, which keeps it short of 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 description coverage is 100% with thorough per-parameter docs (wcag prefixes, level default and AAA interaction, format fallback, minImpact ordering, sourceRoot candidates), so the schema carries the burden. The description spends its words on return values rather than adding parameter meaning, making the baseline 3 correct.

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 precise verb+resource ('Scan a web page for WCAG accessibility issues') and scopes it crisply with 'Works on any URL — deployed sites, localhost, staging', which implicitly separates it from a static-HTML sibling. However, no sibling tool (scan_html, flow_scan, visual_audit) is named, so the differentiation is left for the agent to infer.

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 'works on any URL' scope line implies when the tool is applicable, but there is no explicit when-to-use / when-not guidance and no routing to alternatives among the 19 sibling tools. An agent must infer that a live-URL scan is different from flow_scan or visual_audit on its own.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare readOnly=false, idempotent=false, openWorld=true; the description adds substantially more: async fire-and-return with an audit id, polling via get_audit, durable server-side storage, downloadable report plus Excel workbook, ~5 minute runtime, the page being loaded several times, and the requirement to keep the tunnel open until completion.

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?

Content is dense and mostly earns its place, but the bracketed setup paragraph plus a second 'Free for everyone' statement duplicate the free/account requirement, and the tunnel workflow detail is long. Front-loading of the deliverable framing is good, but there is redundancy.

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 fully carries the return-value burden: it explains the immediate id return, the get_audit polling loop, and the eventual download URLs. For a multi-step async tool with auth and tunnel prerequisites, nothing an agent needs to invoke and follow up 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 description coverage is 100%, so all three parameters are already documented. The description restates includeAgent and tunnel_secret behavior largely in line with the schema, adding only the workflow note that the tunnel must stay open; baseline 3 applies when the schema carries the load.

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+resource (start a FULL accessibility audit deliverable for a URL) and immediately scopes it as a persistent, timestamped artifact rather than an inline scan. It explicitly names the sibling it is not (scan_page + verify_fix for iterating on code), so an agent can differentiate without opening schemas.

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?

Gives explicit when-to-use ('durable artifact to attach as evidence for a compliance officer or legal response') and when-not-to-use with the named alternative (scan_page + verify_fix). It also states prerequisites: free account, hosted MCP connection, and the tunnel workflow for local dev servers.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint/idempotentHint/openWorldHint, but the description adds substantial behavior beyond them: verification fails closed, a blocked page (bot-challenge/HTTP error) yields unverified rather than a pass, and a no-match selector returns verified: false with reason "not-found". These are exactly the failure-mode details an agent needs to interpret results correctly.

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?

Front-loaded with the core action and the loop it closes, and every sentence carries operational content. It is denser and longer than strictly necessary — several em-dash clauses and edge cases could be tightened — but nothing is 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?

For a read-only verification tool with no output schema, the description still conveys the return shape (verified boolean plus a reason value), the engine limitation (DOM engines only), and the recovery path for a missing selector. Nothing an agent needs to call it or interpret it 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%, but the description still adds meaning: the WCAG/axe parameter narrows verification to one criterion while omitting it requires the element to be clean of ALL violations, and the selector should be the one carried over from the original scan_page issue. It also advises reusing the same viewport the issue was found at, which the schema does not state.

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 precise verb+resource: re-scan a specific element after a fix and confirm the violation is gone, producing a machine-checked verified: true|false. It explicitly scopes itself against siblings ("closes the loop that find-only tools leave open") and excludes visual_audit findings, so an agent can distinguish it from scan_page, visual_audit, and diff_scan 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?

Gives explicit preconditions (edit the code and serve it — deployed, staging, or localhost), the exact moment to call it, and the workflow it belongs to (scan_page → generate_ai_fix → verify_fix). It also names what is out of scope (visual_audit findings) and what to do when the selector matches nothing (pass the current selector or re-run scan_page), which is unusually complete routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare openWorldHint=true and non-readOnly/non-idempotent, but the description adds material context annotations cannot convey: it requires authentication, has fair-use rate limits, and drives a headless browser. It does not explain why the tool is marked non-readOnly, leaving that annotation unqualified.

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?

The technical core is strong, but the text is padded with repeated onboarding/billing content — the WebAbility account requirement, MCP URL, and sign-in instructions appear twice across bracketed and closing sentences. The first sentence is entirely setup boilerplate rather than purpose, hurting front-loading.

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?

Prerequisites, capabilities, and sibling relationship are covered well, but with no output schema the description omits what the audit actually returns (report shape, issue list, severity). For an audit tool this leaves the agent guessing about the response.

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 description coverage is 100%, so all four parameters (url, fullPage, viewport, brandColors) are already documented in the schema. The description adds only the redundant note that it 'takes a URL' and gives no extra syntax or format guidance, so the baseline of 3 applies.

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 states a specific verb+resource ('Pixel-level accessibility audit using Claude vision') and enumerates the concrete issue classes it detects with WCAG references. It distinguishes itself from DOM scanners and from the sibling scan_page, so an agent can tell exactly what this tool does without opening 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?

It explicitly routes the agent relative to an alternative: 'Complements scan_page — run both for full coverage,' and frames its niche as 'issues that DOM scanners miss.' Prerequisites (account required, not runnable in Lite) are also stated. It lacks any explicit when-not-to-use guidance, so it falls just short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.