agentleFS
Sign inStart free
fusebase-devfusebase-dev/fusebase-flow/.agents/skills/mcp-gate-debug/SKILL.md

mcp-gate-debug

Skill9 starsChanged 27 days agoMITView on GitHub

After Fusebase Gate MCP tool sessions, produce a concise debug-oriented summary — what went smoothly, what did not, and actionable improvements to fusebase-gate skills, prompts, or MCP server behavior. Prioritize isolated SQL/NoSQL store flows.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    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. Very specific:

    After you finish (or pause) a coherent batch of Gate MCP tool_call operations — especially:

    line 18
  2. Distinctive:

    If the user only asked a one-line question with no tools, skip this summary unless they ask for it.

    line 23
  3. Distinctive:

    When isolated-store tools were used, before proposing doc changes, re-read if needed:

    line 38
  4. Distinctive:

    Do not dump raw tool JSON.

    line 53
  5. Very specific:

    If the fix belongs in this repo’s project-template only, say so.

    line 54

≈ “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 Gate debug summary

Purpose

Close the loop on Gate MCP work: turn friction into specific improvement ideas for:

  • .claude/skills/fusebase-gate/ (especially SKILL.md and references/*.md)
  • MCP prompts / tool descriptions (upstream server)
  • Local app code only when the issue is clearly client-side misuse

When to run

After you finish (or pause) a coherent batch of Gate MCP tool_call operations — especially:

  • Isolated stores — listIsolatedStores, createIsolatedStore, initIsolatedStoreStage, migration status/apply, row CRUD, queryIsolatedStoreSql, NoSQL equivalents
  • Org users, tokens, permissions, health/bootstrap when those sessions were rough

If the user only asked a one-line question with no tools, skip this summary unless they ask for it.

Output format (in the conversation)

Keep it scannable:

  1. Went well — bullets (tools, prompts, docs that helped).
  2. Friction — wrong assumptions, missing context, repeated tools_describe, unclear errors, wrong orgId/storeId/stage, migration/drift confusion, permission surprises.
  3. Improvements — ordered by impact. For each idea, prefer:
    • Target: file path (e.g. fusebase-gate/references/isolated-sql-integrator-troubleshooting.md) or “MCP tool X description / prompt Y”
    • Change: one sentence
    • Why: link to the failure you saw

Isolated stores first

When isolated-store tools were used, before proposing doc changes, re-read if needed:

  • .claude/skills/fusebase-gate/references/isolated-sql.md
  • .claude/skills/fusebase-gate/references/isolated-sql-migration-discipline.md
  • .claude/skills/fusebase-gate/references/isolated-sql-integrator-troubleshooting.md
  • .claude/skills/fusebase-gate/references/isolated-nosql.md (if NoSQL)

The summary should call out:

  • Missing playbook steps (status before apply, dryRun, checksum/version discipline).
  • Confusion between dev and prod stages.
  • Whether structuredIssues / drift messaging was understandable.

Boundaries

  • Do not dump raw tool JSON; summarize patterns.
  • If the fix belongs in this repo’s project-template only, say so; if it belongs to the Gate MCP service or platform, label it as upstream.

Flag

This skill is copied into the project only when the global CLI flag mcp-gate-debug is enabled (fusebase config set-flag mcp-gate-debug then fusebase update --skip-mcp --skip-deps --skip-cli-update --skip-commit).

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