agentleFS
Sign inStart free
MCamnerMCamner/mq-image-analyze/skills/mcp-tools-maintainer/SKILL.md

mcp-tools-maintainer

Skill1 starsChanged 22 days agoMITView on GitHub

Use when changing mq-image-analyze MCP server, MCP tool schemas, tool safety, read-only behavior, or mq-mcp/mq-agent integration.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill for MCP behavior and tool contracts.

    line 8
  2. Distinctive:

    All MCP tools are read-only by default.

    line 36
  3. Very specific:

    Keep the safety classification aligned with docs/tool-safety.md.

    line 40
  4. Distinctive:

    do not hide uncertain detections without reporting why

    line 50
  5. Distinctive:

    When changing an MCP tool:

    line 54
  6. Very specific:

    Update schema/docs in docs/mcp-tools.md.

    line 57
  7. Distinctive:

    Update JSON contract docs if output changes.

    line 58
  8. Very specific:

    Add or update tests/test_mcp_tools.py.

    line 59
  9. Distinctive:

    Check mq-agent and mq-mcp integration docs when relevant.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

MCP Tools Maintainer

Use this skill for MCP behavior and tool contracts.

Evals

Should trigger

  • "add an MCP tool to the image server"
  • "tighten MCP tool read-only safety"
  • "fix mq-mcp integration of the image tools"
  • "the MCP tool schema is wrong"

Should not trigger

  • "change a CLI flag" → use cli-maintainer
  • "change the JSON result schema" → use json-contract-maintainer
  • "set up a model backend" → use model-setup-maintainer

Core Files

  • mq_image_analyze/mcp/server.py
  • docs/mcp-tools.md
  • docs/tool-safety.md
  • docs/integration.md
  • tests/test_mcp_tools.py
  • tests/test_json_schema.py

Safety Contract

All MCP tools are read-only by default. They inspect image files and return structured analysis. They must not write, delete, mutate, or suppress detector output.

Keep the safety classification aligned with docs/tool-safety.md.

No Silent Omission Rule

For exhaustive mode:

  • preserve every detection
  • include confidence, bbox, area percent, and source model
  • include content flags
  • include limitations
  • do not hide uncertain detections without reporting why

Tool Contract Changes

When changing an MCP tool:

  1. Update implementation.
  2. Update schema/docs in docs/mcp-tools.md.
  3. Update JSON contract docs if output changes.
  4. Add or update tests/test_mcp_tools.py.
  5. Check mq-agent and mq-mcp integration docs when relevant.

Verification

python -m pytest tests/test_mcp_tools.py tests/test_json_schema.py -q
python -m compileall mq_image_analyze/mcp -q

For full safety:

bash scripts/validate.sh

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