Mcp Tool Docs Stinger logo

Mcp Tool Docs Stinger

Organization
legioncodeinc
mcp-tool-docs-stinger

Documents MCP tools, TypeScript APIs, and CLIs honestly: schema, output, side effects, examples, doc-to-code sync. Use when writing or auditing tool/API/CLI docs.

Overview

Publisherlegioncodeinc
Repositoryvibe-coding-tools
Skill namemcp-tool-docs-stinger
Stars
94
Forks
44
Bundled files
18
LicenseAGPL-3.0-or-later
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 18 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by legioncodeinc on GitHub. Read the source before you install it.

Installation

Install the Mcp Tool Docs Stinger AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/legioncodeinc/vibe-coding-tools/tree/main/plugins/documentation-tools/skills/mcp-tool-docs-stinger
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/legioncodeinc/vibe-coding-tools.git /tmp/vibe-coding-tools
mkdir -p .claude/skills
cp -r /tmp/vibe-coding-tools/plugins/documentation-tools/skills/mcp-tool-docs-stinger .claude/skills/mcp-tool-docs-stinger
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Mcp Tool Docs Stinger in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Mcp Tool Docs Stinger on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Mcp Tool Docs Stinger is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

mcp-tool-docs-stinger

Procedural arsenal for mcp-tool-docs-wasp-drone, the Wasp Nest's tool/API/CLI documentation specialist. This stinger encodes how to document any tool, API, and CLI surface honestly: schema-selected tools (MCP tools first and foremost - name, purpose, input schema, output shape, side effects, annotations, examples), the TypeScript public API rendered with a generator (TypeDoc, and API Extractor where a reviewable public-API contract is needed), a CLI's command surface, doc-to-code sync discipline, and changelog discipline tied to a released artifact's real version.

Hivemind (@deeplake/hivemind) - the MCP tools exposed by src/mcp/server.ts plus the OpenClaw goal/KPI contracts, its TypeScript public API, the hivemind CLI, and its sync-versions-driven changelog - remains fully covered throughout as a worked example. The guides teach the general practice first and point to the Wasp Nestmind case for a real, end-to-end illustration.

When this stinger applies

Load this stinger when mcp-tool-docs-wasp-drone is invoked. Typical triggers:

  • "Document these MCP tools."
  • "Is the description on this tool honest? Does it match the code?"
  • "Write the input schema and output shape for this tool."
  • "Generate the TypeScript API reference."
  • "Document this CLI's command surface."
  • "These docs drifted from the code - re-sync them."
  • "Write a changelog entry for this release."
  • "Audit the tool/API/CLI docs."
  • "Document the Wasp Nestmind MCP tools / the hivemind CLI / Hivemind's TypeDoc setup." (the worked-example case)

Do NOT load it for:

  • MCP protocol, transport, or handshake internals (route to mcp-protocol-wasp-drone).
  • Prose-quality review or ghostwriting - Diataxis mode, inverted-pyramid structure, voice/tone, "is this well-written" (route to technical-writing-craft-wasp-drone). This Drone transcribes facts from source honestly; it does not review prose craft.
  • OpenAPI/REST API documentation - Swagger UI / Redoc / Scalar / Mintlify renderer selection, OpenAPI spec example enrichment, REST SDK generation from an OpenAPI spec (route to api-docs-wasp-drone).
  • Docs-site platform and hosting - Docusaurus / Starlight / Mintlify / GitBook selection, docs-as-code CI for a whole site, search setup (route to docs-site-wasp-drone).
  • README authoring as a standalone deliverable (route to readme-writing-wasp-drone).
  • The library/ knowledge convention or general knowledge capture docs (route to library-wasp-drone / knowledge-wasp-drone).
  • Deeplake dataset schema design (route to vector-store-wasp-drone).

First action when this stinger is loaded

Read these in order before doing anything else:

  1. guides/00-principles.md - doc honesty, the five quality gates, when to route elsewhere, and the core invariants (general, not product-specific).
  2. guides/01-mcp-tool-docs.md - how to document any schema-selected tool from its real registration and handler. Read this before documenting any tool.
  3. research/distilled-mcp-tool-docs.md - the current synthesis: honest MCP tool docs (including annotations), TypeScript API reference generation beyond TypeDoc, CLI documentation conventions, doc-to-code sync tooling, and changelog automation.

Then walk the remaining guides in task order. Each guide teaches the general practice first, then points to the matching examples/*.md file for a worked Hivemind case. Always read the real source before writing - these docs are honest about the code or they are wrong.

Folder layout

text
mcp-tool-docs-stinger/
├── SKILL.md                          (this file)
├── README.md                         (one-page human overview)
├── guides/
│   ├── 00-principles.md              (doc honesty, five quality gates, scope boundary, cross-links)
│   ├── 01-mcp-tool-docs.md           (documenting any schema-selected tool from its real schema + handler)
│   ├── 02-typedoc.md                 (TypeScript API reference generation: TypeDoc + API Extractor)
│   ├── 03-cli-docs.md                (documenting any CLI's command surface)
│   ├── 04-doc-sync.md                (keeping docs in sync with code; drift detection, hand-rolled or tooled)
│   ├── 05-changelog.md               (changelog discipline tied to a released artifact's version)
│   └── 06-done-checklist.md          (10-point validation before docs ship)
├── examples/                         (worked examples - all Hivemind-specific, clearly labeled)
│   ├── hivemind-search-tool-doc.md   (full worked doc for the hivemind_search MCP tool)
│   ├── hivemind-cli-reference.md     (CLI reference for install / status / login)
│   ├── typedoc-setup.md              (TypeDoc config + npm script for the TS public API)
│   └── changelog-entry.md            (worked changelog entry for a real version bump)
├── templates/
│   ├── mcp-tool-doc.md               (tool doc template: name/purpose/schema/output/side-effects/examples)
│   ├── cli-command-reference.md      (CLI command reference template)
│   ├── typedoc-json.md               (typedoc.json + package.json script template)
│   ├── docs-sync-workflow.yml        (CI workflow that fails when docs drift from code)
│   └── changelog-entry.md            (changelog entry template tied to a released version)
├── reports/
│   └── README.md                     (how past audit summaries accumulate)
└── research/
    ├── distilled-mcp-tool-docs.md    (current synthesis, 2026-08-14 - read this first)
    ├── research-plan.md              (original Hivemind-anchored plan, 2026-06-16 - DO NOT MODIFY)
    ├── research-summary.md           (original Hivemind-anchored summary, 2026-06-16 - DO NOT MODIFY)
    ├── index.md                      (index across both research passes)
    └── external/                     (source notes: 2 from 2026-06-16, 8 new from 2026-08-14)

Surfaces to document

SurfaceSource of truthHow it's documented
Schema-selected tools (MCP, etc.)the server's tool-registration codeName, purpose, input schema, output shape, side effects, annotations (where the protocol defines them), examples
TypeScript public APIexported types + functions in sourceTypeDoc (readable reference), API Extractor (reviewable public-API contract) - pick one or both per guides/02-typedoc.md
CLIthe CLI's dispatch/routing + its own usage/help textCommand reference: usage, flags, side effects, non-interactive path, disambiguation notes
In-repo reference docsREADME, architecture/feature docsKept in sync with code; doc honesty enforced
Changelogthe artifact's single-sourced released versionEntry per released version, hand-written impact-first or generated from Conventional Commits

Worked example (Hivemind): the MCP tools shipped today are hivemind_search, hivemind_read, and hivemind_index (stdio transport, read-only, auth via ~/.deeplake/credentials.json); OpenClaw additionally contracts hivemind_goal_add and hivemind_kpi_add (write tools). The TS public API is rendered with TypeDoc from src/index.ts. The CLI (hivemind install [--only <platforms>] [--skip-auth] [--token <value>], hivemind <agent> install, hivemind uninstall, hivemind login, hivemind status, hivemind update [--dry-run], plus goal/kpi/context/graph/dashboard/rules/skillify/embeddings <sub>) is documented from src/cli/index.ts routing - never from memory. The changelog tracks @deeplake/hivemind, single-sourced via scripts/sync-versions.mjs. See guides/01-mcp-tool-docs.md through guides/05-changelog.md for the general procedure each of these demonstrates, and examples/*.md for the worked docs themselves.

Cross-links (adjacent Drones - route there instead of duplicating)

This stinger owns the source-derived reference-docs layer for tool, API, and CLI surfaces. Three adjacent stingers own territory this one does not - cross-link rather than re-teaching their material:

  • technical-writing-craft-stinger - the craft of writing well: Diataxis mode classification, inverted-pyramid structure, code-example discipline, voice/tone, ghostwriting, docs-as-code prose review. Use it when the question is "is this well-written," not "does this doc match the code."
  • api-docs-stinger - OpenAPI/REST API documentation specifically: Swagger UI / Redoc / Scalar / Mintlify / Stoplight / Bump.sh tool selection, OpenAPI spec example enrichment, REST SDK generation (openapi-generator-cli, Fern, Speakeasy). Use it for a REST API's OpenAPI spec and its renderer; use this skill for the TypeScript SDK's exported symbols and any MCP/CLI surface.
  • docs-site-stinger - documentation-site platform selection and hosting (Docusaurus, Starlight, Mintlify, GitBook, MkDocs Material, Nextra, Fern), docs-as-code CI for a whole site, search setup. This skill's generated API reference is often published inside one of those sites - picking and running the site itself belongs there.

Critical directives (lifted from the Command Brief)

These are non-negotiables. Full justification in guides/00-principles.md.

  • Read the source before writing a single line. A tool doc that does not match src/mcp/server.ts is a bug, not documentation.
  • Tool descriptions and schemas must match real behavior. The zod inputSchema, the output content shape, and the side effects are facts, not prose. Honest or wrong, no middle.
  • Every MCP tool doc carries six parts: name, purpose, input schema (from zod), output shape, side effects, and at least one example.
  • TypeDoc renders from the TS types, not hand-written prose. Fix the doc comment in the source; never fork the truth into a separate file.
  • The changelog is tied to the npm version. scripts/sync-versions.mjs single-sources the version; the changelog tracks @deeplake/hivemind releases, not arbitrary dates.
  • Do not scope-creep into protocol internals or README authoring. Route to mcp-protocol-wasp-drone / readme-writing-wasp-drone.
  • This is a general documentation practice, not a single-product skill. The Wasp Nestmind-specific facts above still hold exactly as written for Hivemind; when documenting any other project's tools, API, or CLI, apply the same shape (six-part tool docs, generated API references, source-derived CLI references, single-sourced changelog versions) to that project's real source instead.
  • Set and check MCP tool annotations, not just prose side effects. Where a server defines readOnlyHint / destructiveHint / idempotentHint / openWorldHint, the tool doc records the real values and flags any contradiction with the prose side-effect claim; where none are set, say so and note the pessimistic default a client will assume.
  • Do not scope-creep into OpenAPI/REST docs, docs-site platform work, or prose-craft review. Route to api-docs-wasp-drone, docs-site-wasp-drone, or technical-writing-craft-wasp-drone respectively - see Cross-links above.

Forged by stinger-forge from mcp-tool-docs-wasp-drone-command-brief.md and research/. Broadened to general tool/API/CLI documentation practice 2026-08-14, Hivemind material preserved as worked examples. Part of The Wasp Nest, curated by Mario Aldayuz a.k.a @thenotoriousllama.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Mcp Tool Docs Stinger AI skill do?

Documents MCP tools, TypeScript APIs, and CLIs honestly: schema, output, side effects, examples, doc-to-code sync. Use when writing or auditing tool/API/CLI docs.

Why use Mcp Tool Docs Stinger on TypingMind?

Because you install it once and use it with any model. Mcp Tool Docs Stinger is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Mcp Tool Docs Stinger in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/legioncodeinc/vibe-coding-tools/tree/main/plugins/documentation-tools/skills/mcp-tool-docs-stinger. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Mcp Tool Docs Stinger?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Mcp Tool Docs Stinger?

As many as you like. As long as a model supports skills, you can use Mcp Tool Docs Stinger with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Mcp Tool Docs Stinger AI skill free?

Yes. It is published on GitHub by legioncodeinc under the AGPL-3.0-or-later license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇