docs-mcp

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

Goes well beyond the readOnly/non-destructive annotations by disclosing non-obvious mechanics: 'platform' filters results AFTER retrieval and can cut a 15-section response to 1-2, 'version' is applied DURING retrieval and mainly prevents per-version duplication, and version-neutral docs (Cloud, Connect, API reference) are always included. These are exactly the facts an agent needs to predict result size.

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 core purpose and result contract are front-loaded, and the sentences about 'platform' and 'version' each carry decisions the agent must make. It is long and restates version-neutral doc behavior that also appears in the schema, so a few sentences are not strictly additive.

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 supplies the return contract (sections with url plus markdown content, ordered by relevance) and covers defaults for every optional parameter. An agent has everything needed to call it correctly and interpret the result.

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 80%, so most parameter meaning is already in the schema; the description still adds value with the concrete 15-to-1/2 shrink example for 'platform' and the explicit statement that omitting 'version' searches the latest release. The added detail largely mirrors the schema text, keeping it out of 5 territory.

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 opening sentence names a specific verb and resource ('Search the official Redpanda documentation') and states the return shape ('most relevant sections... url and its actual content in markdown'). This is clearly distinguishable from the sibling API-reference tools, which target a narrower corpus.

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?

'Use this tool for all queries that require Redpanda knowledge' gives an explicit trigger, and the description then walks through the tradeoff of supplying vs omitting 'platform' and 'version'. It never names an alternative tool or a when-not-to-use case, so it stops 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 readOnlyHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds genuinely useful context beyond that: a hard batch limit of 10 pages per request and the constraint that URLs must live under docs.redpanda.com/api/doc/*. It does not describe pagination or response format, but the domain and batch constraints are the material extras here.

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?

Four short sentences, each carrying distinct information: what it returns, when to call it, input sourcing, and hard limits. The core purpose is front-loaded and nothing is padded.

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 explaining returns and does so ('complete endpoint details including parameters, request/response schemas, and examples'). The only minor gap is the absence of any note on what happens with invalid or partially-failing URLs when multiple are batched.

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 both parameters including the optional 'context' field are documented in the schema, setting a floor of 3. The description adds meaning beyond the schema by naming the permitted URL domain and reiterating where the URLs should originate, which constrains what a valid value actually is.

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 ('Retrieve full content of API reference pages by URL') and immediately distinguishes itself from siblings by positioning as the follow-up to list_api_reference_pages and search_api_reference. An agent can tell it fetches full page bodies rather than listing or searching.

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 states when to use it ('Use this after finding pages via list_api_reference_pages or search_api_reference') and how to source inputs ('Pass the URLs from those results directly'). It routes the agent away from using it for discovery, which is the sibling tools' job.

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, destructiveHint=false, openWorldHint=false, so safety is covered. The description adds genuine value not in the annotations: it discloses the returned fields (URL, title, type, description), which matters because there is no output schema. It does not mention pagination or result-size limits.

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 purpose and return fields, then a scannable scoping block. The api list partially overlaps the schema enum, but each line adds semantic content rather than repeating it verbatim, so the length is largely earned.

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, so the description correctly carries the return-format burden, and it covers the primary scoping decision. Minor gaps remain (pagination, behavior of the url parameter beyond the schema note, relationship to search_api_reference), but nothing critical to 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 baseline is 3, but the description goes beyond the schema's bare value list by explaining what each api value actually contains (e.g., admin = brokers/partitions/configs/users, http-proxy = produce/consume/offsets). That semantic mapping is real added meaning an agent can use without opening 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 and resource ("List pages in Redpanda API reference documentation") and immediately specifies the payload (endpoints, schemas, topic pages with URL, title, type, description). This distinguishes it cleanly from the doc-Q&A sibling ask_redpanda_question.

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 ("Use this to browse API structure") and names one alternative with the condition that selects it ("For general Redpanda docs, use ask_redpanda_question instead"). It does not route the agent among the closer siblings search_api_reference and get_api_reference_content, so the guidance is clear but not complete.

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, destructiveHint=false, and openWorldHint=false, so safety is covered. The description adds genuine context beyond that: a hard result cap of 20 and the return field shape (URL, title, text excerpts), which matters since no output schema exists. It stops short of describing pagination, truncation behavior when more than 20 match, or ranking, so it is strong but not complete.

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 purpose and return shape, then organized into clearly labeled SCOPING and WHEN TO USE sections. It is on the long side, and the example mappings partially restate the scoping bullets, but each line is scannable and actionable rather than 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?

For a 4-parameter read-only search tool with no output schema, the description covers purpose, all scope options, selection heuristics, the result cap and result fields, and the sibling alternative. An agent has everything needed to call it correctly on the first try.

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 would be 3. The description exceeds that by explaining the semantic meaning of each `api` scope (which resources live in admin vs cloud-controlplane vs cloud-dataplane) rather than merely restating the enum values, letting the agent choose a scope correctly. `type` and `context` are left to the schema, which is acceptable at full coverage.

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 ('Search Redpanda API reference documentation by keyword') and immediately describes the result shape ('up to 20 matching endpoints, schemas, or topics with URL, title, and text excerpts'). It also names the sibling it is not (ask_redpanda_question) for non-API questions, so an agent can distinguish it before 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 SCOPING and WHEN TO USE WHICH sections explicitly enumerate each `api` value, what each covers, and give concrete question-to-scope mappings ('broker endpoints' → admin, 'create topic API' → all). It also states the exclusions via the alternative tool for general Redpanda questions. This is exactly the when/when-not/alternatives guidance the dimension asks for.

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?

Beyond the annotations (openWorldHint=true, destructiveHint=true) the description discloses that a Redpanda Cloud sign-in is required, that the user's work email and company domain are attached, and precisely how to react to an authentication_required refusal (tell the user to reconnect rather than retry). That is exactly the kind of side-effect and auth context annotations alone 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?

The purpose is front-loaded in the opening sentence and each subsequent sentence carries distinct load (consent, data disclosure, auth requirement, failure handling). It is on the long side, but the length is justified by the compliance and auth behavior it documents rather than 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?

With no output schema, the description still covers what the agent needs: trigger conditions, consent requirement, what data leaves the client, and how to handle auth failure. Nothing material 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, but the description adds real guidance: it instructs the agent to summarize feedback clearly and to include the documentation page URL or context when known, which maps directly to filling page_url and context meaningfully. It also implicitly reveals that email/domain are attached server-side without being schema fields.

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 and resource — 'Send feedback about the Redpanda documentation or products directly to the Redpanda team' — and the delivery target is explicit. It is clearly distinguishable from the sibling read/search tools (ask_redpanda_question, search_api_reference, etc.), which cannot send anything.

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 enumerates the exact triggering conditions (bug, documentation gap, incorrect or missing info, user frustration), mandates asking for consent first, and explicitly says 'never submit feedback without their consent.' It also gives a when-not-to-retry rule tied to the authentication_required refusal, 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.