Build docs/mcp-tools.md from the active MCP servers and available tool functions.
Installation
Details
Usage
After installing, this skill will be available to your AI coding assistant.
Verify installation:
npx agent-skills-cli listSkill Instructions
name: mcp-tools-catalog description: Build docs/mcp-tools.md from the active MCP servers and available tool functions. metadata: short-description: Generate MCP tool catalog
MCP Tools Catalog
Create a project-local catalog of MCP tools at docs/mcp-tools.md.
Core rules
- Use the runtime MCP tool list and enabled server list to generate the catalog.
- Draft in chat first; ask for one confirmation before writing any files.
- Do not edit other files.
- Exception: if
AGENTS.mdlacks a reference todocs/mcp-tools.md, include that update under the same confirmation.
Steps
- Run:
codex mcp list --jsonto get enabled server names. - Use prompt-engineering to list the tool functions currently available right now using this exact prompt:
list the functions that is CURRENTLY avalible to you **RIGHT NOW**, starting with the prefix `functions.mcp__`
after each function, provide a clear, concise, and informative overview of what each function is intended for
- Map tool functions to
server:toolnames (e.g.,playwright:browser_click). - Filter out any tools whose server is not enabled.
- Build
docs/mcp-tools.mdusingassets/mcp-tools-template.md. - If
AGENTS.mdlacks a reference todocs/mcp-tools.md, include that change in the draft summary. - Ask: "Reply CONFIRM to write docs/mcp-tools.md and update AGENTS.md (if needed)."
- On confirmation, write the file and update
AGENTS.mdif missing the reference.
Output requirements
- Use
server:toolnaming per line. - Include a short description for each tool.
- Include a list of enabled servers at the top.
More by appautomaton
View allDraft a repo-local plan using the plan skill template and optionally save it.
Write, configure, and run unit + integration tests using project-independent best practices; detect tooling; triage failures/flakes; keep a compact Testing State Capsule.
Generate a project-specific AGENTS.md from a user goal, then confirm before overwriting.
Use the Claude Code CLI to consult Claude and delegate coding tasks for prototyping, debugging, and code review. Supports multi-turn sessions via SESSION_ID. Optimized for low-token, file/line-based handoff.
