claude

mega-loop

Grade your traces against MEGA Loop's readiness contract and fix the instrumentation, then browse the bugs MEGA Loop diagnosed from those traces, autofix them, and get a draft PR — from inside Claude Code. The engine finds the bug and judges the fix; your session runs the code.

6 on GitHubby Mind AIv0.2.0Proprietaryupdated 3 days ago
/

Plugin mega-loop from marketplace mega-loop in mega-edo/mega-loop-skill. Commands read from the repo's manifest — 2 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (7)

  • connect

    Pick which mega-loop project this session works on. Use when the user says "connect", "which project", "switch project", or before listing bugs / fixing when no project is chosen yet. Lists the user's projects and remembers the one they pick for later commands.

  • diagnose

    Read what mega-loop diagnosed from production traces — list the open bugs for your project, and explain one bug's cause + the exact file:line. Use when the user asks "what's broken / what should I fix / where / why" about a mega-loop project, or says "bugs" / "explain". Read-only: it reports, it nev

  • fix

    Fix a bug that mega-loop diagnosed from production traces — from inside this session. Use when the user asks to fix/autofix a mega-loop bug, mentions a bug id or bug title from the mega-loop dashboard, or says "fix" / "autofix". One verb; the SERVER decides who fixes: project bound to a GitHub repo

  • refine

    Apply maintainer review feedback to an open auto-fix PR/MR — a REVISION, not a fresh fix. Use when the user says "refine", "apply the review comments", "address the PR feedback", or after a maintainer reviews a mega-loop fix PR. Reads the PR's review comments (GitHub / GitLab / Bitbucket), makes the

  • trace-analyze

    Grade an AI agent's traces against the readiness checks MEGA Loop runs internally, plus this plugin's own checks of what its detection reads, and produce a report with a fix for every finding — read-only, no edits. Use when the user's traces are missing, fragmented or ignored, when MEGA Loop reports

  • trace-fix

    Fix an AI agent's instrumentation so its traces become usable by MEGA Loop — one clean trace per request, OpenInference-compliant — then re-run the validator to prove it. Edits the user's code. Use after trace-analyze finds problems, or when the user asks to make their traces pass MEGA Loop's readin

  • trace-gen

    Add tracing to an agent that emits none, then prove the first traces are readable by running the app and grading what came out. Use when a repository has no instrumentation at all, when the user asks how to start tracing, or when MEGA Loop has no traces to read because none are being sent. Use trace

Slash commands (6)

  • /mega-loop:bugs

    List the open bugs mega-loop diagnosed for the active project

  • /mega-loop:disconnect

    Disconnect this repo from its saved mega-loop project (stop auto-reconnecting)

  • /mega-loop:explain

    Explain one mega-loop bug — its cause and the exact file:line

  • /mega-loop:groups

    List the project's bugs grouped into packaging units (the shape a fix ships)

  • /mega-loop:projects

    List your mega-loop projects (read-only; use /mega-loop:connect to pick one)

  • /mega-loop:status

    Show mega-loop status — setup/auth, the connected project, all your projects, and any fix in flight

MCP servers (1)

  • mega-loop

    connects to: ${user_config.base_url}/companion/mcp

Hooks (2)

  • SessionStart

    ${CLAUDE_PLUGIN_ROOT}/hooks/restore-connection.sh

  • Stop

    ${CLAUDE_PLUGIN_ROOT}/hooks/gate-on-stop.sh

View source on GitHub

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

CMEM Pro

claude

CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents

3 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Structured logging, distributed tracing, and metrics — correlate agent swarm activity with application telemetry

2 skills · 1 command · 1 agent

ruvnet/ruflo 81,589

Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer

8 skills · 1 MCP server

ChromeDevTools/chrome-devtools-mcp 55,181

Install the "AAS Observability & Incident Response" editorial skill bundle for Claude Code.

12 skills

sickn33/agentic-awesome-skills 49,620

gitnexus

claudecodex

Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.

12 skills · 1 MCP server · hooks

abhigyanpatwari/GitNexus 48,482

Maintain mega-loop?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted