The documentation, as a tool your agent can call: 950+ AI-dev guides. Search + fetch tools.
Scored across 2 tools
search and fetch have clearly distinct roles — one discovers via snippets and IDs, the other retrieves full text by ID. The descriptions explicitly state the handoff, so there is no overlap or misselection risk.
Both tool names are simple, lowercase verbs (search, fetch) with no mixed conventions or styles. The pattern is predictable and consistent throughout the server.
Only two tools are provided. While each is necessary for a read-only documentation workflow, the set is thin and lacks complementary operations like browsing or listing categories, which the rubric marks as borderline.
search and fetch cover the core discover-and-read lifecycle for documentation. Minor gaps exist: no browsable index, no pagination beyond 10 results, and no metadata-only lookup, but these are workable for the stated purpose.
Retrieve the complete markdown of one documentation article by the id returned from search (for example en/claude-code/advanced-techniques/hooks-automation). The text is returned in full; metadata.gated only reports whether the article sits behind the paywall on the web. An unknown id is an error — call search first.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Article id from a `search` result. | |
| llm_model | Yes | The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" — never guess. |
| Name | Required | Description |
|---|---|---|
| id | Yes | |
| url | Yes | |
| text | Yes | |
| title | Yes | |
| metadata | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered. The description adds valuable behavioral context beyond that: the full markdown is returned, `metadata.gated` only reflects web paywall status, and unknown ids produce an error. This is a clear improvement over the structured metadata alone.
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 every sentence adds information: the retrieval behavior, the paywall semantics, and the required prerequisite. There is no filler or redundant repetition of schema/annotation content.
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?
Given that an output schema exists and the input schema covers all parameters at 100%, the description does not need to explain return shape or parameter formats. It covers the necessary behavioral aspects—full content, error condition, and the relationship to `search`—leaving no critical gap for correct invocation.
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 adds extra meaning for the `id` parameter by giving a concrete example of the expected format and reinforcing that it must come from a `search` result. The `llm_model` parameter is already fully documented in the schema, so no further description contribution is needed there.
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 title and description clearly state the resource (a documentation article) and the specific operation (retrieve complete markdown by id). It explicitly ties the input to values returned by `search`, which distinguishes it from its sibling tool without ambiguity.
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 explicitly instructs the agent to call `search` first and warns that an unknown id is an error. This gives clear when-to-use guidance and prevents misuse, even though it does not enumerate every non-use case explicitly.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the AI Developer Toolkit documentation: 1100+ guides on Cursor, Claude Code and OpenAI Codex, covering setup, agent workflows, hooks, MCP, testing, CI and deployment, in English and Polish. Returns at most 10 ranked results, each with a short snippet rather than the article text; an empty list means the corpus has nothing on the topic. Pass a result id to fetch for the full text.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | Natural-language query or keywords. Polish queries return Polish articles. | |
| llm_model | Yes | The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" — never guess. |
| Name | Required | Description |
|---|---|---|
| results | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish readOnlyHint=true and openWorldHint=false, so safety is covered. The description adds real behavioral context: a hard cap of 10 ranked results, snippet-only returns, and the meaning of an empty list as 'nothing in corpus' — details not derivable from annotations or 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?
Three tight sentences: corpus scope first, return behavior second, handoff to fetch last. No redundant restatement of the title or annotations, and every clause carries information an agent needs.
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?
Even though an output schema exists (so return values need not be explained), the description usefully summarizes the result shape, cap, empty-list semantics, and the follow-up call. Nothing material 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 description coverage is 100%, so both parameters are already documented, including the Polish-query behavior and the 'never guess' rule for llm_model. The description adds no further syntax or format detail, making 3 the correct baseline.
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 (Search) and resource (AI Developer Toolkit documentation) plus the corpus scope: 1100+ guides across named tools, languages, and topics. It also distinguishes itself from the sibling 'fetch' by explaining that search returns snippets and fetch returns full text.
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 clearly implies the search-then-fetch workflow and names the sibling to use for full article text, giving an agent a routing rule. It does not state when *not* to search (e.g. for a known id), so it stops short of full when/when-not guidance.
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.
fetch2 fields changedInput schema / properties / llm_modelAdded value: +{
+ "description": "The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. \"claude-opus-4-8\", \"gpt-5.2\"). Used for analytics only. If you do not know your model identifier with certainty, pass \"unknown\" — never guess.",
+ "type": "string"
+}Input schema / requiredPrevious value: -[
- "id"
-]New value: +[
+ "id",
+ "llm_model"
+]search2 fields changedInput schema / properties / llm_modelAdded value: +{
+ "description": "The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. \"claude-opus-4-8\", \"gpt-5.2\"). Used for analytics only. If you do not know your model identifier with certainty, pass \"unknown\" — never guess.",
+ "type": "string"
+}Input schema / requiredPrevious value: -[
- "query"
-]New value: +[
+ "query",
+ "llm_model"
+]fetchsearchDocs Q&A: search 169 data and AI guides, fetch any page as markdown. Read-only, keyless.
Tailor Platform for AI assistants: search, list and read the platform documentation.
Gemmein docs, API reference, error explainers and a live integration check for coding agents.
Add one secure layer between your agents and this server.