agentleFS
Sign inStart free
dylanroscoverdylanroscover/Embody/.claude/skills/add-mcp-tool/SKILL.md

add-mcp-tool

Skill219 starsChanged 2 months agoMITView on GitHub

Add a new MCP tool to the Envoy server (Embody development)

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat these as API contracts - changes break client integrations

    line 13
  2. Very specific:

    Update the /mcp-tools-reference skill in .claude/skills/mcp-tools-reference/SKILL.md

    line 26
  3. Distinctive:

    Update the corresponding template DAT if it exists

    line 27
  4. Distinctive:

    Add to the root CLAUDE.md if the tool is significant

    line 30
  5. Very specific:

    Update text_claude.md template DAT for user projects

    line 31
  6. Distinctive:

    Verify the tool appears in the tool list

    line 34
  7. Distinctive:

    Test with valid and invalid inputs

    line 35

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Add MCP Tool to Envoy

Steps for adding a new MCP tool:

  1. Add the tool function inside _register_tools() in EnvoyExt.py
    • Function signature and docstring define the MCP schema (parameter names, types, descriptions)
    • Treat these as API contracts - changes break client integrations
    • The tool function queues the operation for main-thread execution
  2. Add a handler case in _onRefresh() for the TD operation
    • This is where the actual TouchDesigner operations execute (on the main thread)
    • Wrap TD operations in try/except, return {'error': str(e)} dicts on failure
    • Validate all inputs before passing to TD
  3. Update EXPECTED_ENVOY_TOOLS in dev/embody/unit_tests/test_agent_contract.py
    • The Tier-1 agent contract test fails on inventory drift in either direction -- same commit as the tool change
    • Tools register on the MCPServer instance at Envoy START -- restart Envoy before believing a mismatch
  4. Update the MCP tools reference
    • Update the /mcp-tools-reference skill in .claude/skills/mcp-tools-reference/SKILL.md
    • Update the corresponding template DAT if it exists
  5. Update documentation
    • Add to the root CLAUDE.md if the tool is significant
    • Update text_claude.md template DAT for user projects
  6. Test via MCP Inspector or Claude Code
    • Verify the tool appears in the tool list
    • Test with valid and invalid inputs
    • Check error handling

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free