kick-ass-loop-engineer
CLAUDE.md0 starsChanged 7 days agoMITView on GitHub
Model-agnostic, multi-agent build/review loop. A local builder model (Ollama/Kimi) writes code inside the engine pipeline; the Claude Code session is the dispatcher — it loops on loop-engineer next and executes the envelopes it emits.
In this repositoryOne other file it gives its agents.
- loop-engineerSkill
The file
We split this CLAUDE.md into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Write full files, docstrings, exception handling.
- Very specific: line 20
Never relax the security guardrails (see the skill's references/security.md).
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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.
Files
kickAssLoopEngineer
Model-agnostic, multi-agent build/review loop. A local builder model (Ollama/Kimi)
writes code inside the engine pipeline; the Claude Code session is the dispatcher —
it loops on loop-engineer next and executes the envelopes it emits.
Invoking
/loop-engineer <objective> runs the loop via .claude/skills/loop-engineer/SKILL.md.
Key commands
loop-engineer refine --build "<idea>" --done-when "<criteria>" --workspace <dir>— assess the idea and write SPEC.md/GOAL.md (exit 3 + gap JSON when not ready).loop-engineer next --workspace <dir> [--available <csv>] [--objective "<goal>"]— emit ONE dispatcher envelope (invoke_skill | invoke_agent | run_engine | ask_user | terminal); always exits 0, state lives in the envelope.loop-engineer run— full verified pipeline (decompose → ensemble in worktrees → gates+proof → cross-model review → promote), JSON RunOutcome out.loop-engineer verify --gate <unit|security|data_leak|performance|smoke> --cmd "<allowlisted>" [--prove] --workspace <git-dir>— safe, gate-based verification that emits a JSON evidence record.loop-engineer build— one guarded builder round (low-level manual tool; the pipeline runs this internally).
Conventions
- Generated work lives under the chosen workspace; run artifacts under
.loop-engineer/(pipeline.json cursor, next.json retry ledger, stage artifacts like plan.md/review.md). - Write full files, docstrings, exception handling; no noise comments.
- Never relax the security guardrails (see the skill's references/security.md).
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
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.