tomgould/mullet-public
Gated unattended coding jobs: Mullet runs a large coding task in a sandboxed clone and a step passes only on its own checks. Delegates to free, local, Kimi, OpenRouter or Claude models.
ARCHETYPE
Minimalist
Lean instructions, few tools, tiny context tax. Lets the model think.
Copy this rig
# review before running: this installs third-party code
$ npx degit tomgould/mullet-public/.claude ./rig-mullet-public # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ curl -fsSL --create-dirs -o .claude/agents/patch-writer.md https://raw.githubusercontent.com/tomgould/mullet-public/main/.claude/agents/patch-writer.md $ curl -fsSL --create-dirs -o .claude/agents/test-runner.md https://raw.githubusercontent.com/tomgould/mullet-public/main/.claude/agents/test-runner.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/tomgould/mullet-public/main/.claude/agents/verifier.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Subagents (3)
| patch-writer | Generates unified diff patches for file changes. Use when a file needs editing but you want patch output rather than full file rewrite. Specialist for large file changes (>200 lines). |
| test-runner | Runs the test suite and reports failures. Use after code changes to verify nothing regressed. Returns pass/fail counts and failing test names only. |
| verifier | Verifies a patch or code change is safe to apply. Checks for: syntax errors, schema assumptions, test regressions, security issues. Use before committing delegated changes. |
Similar rigs
Remus3/Legion-Wallpaper
Self-auditing AI image restoration pipeline (super-resolution, LaMa inpainting, face repair, metric-plus-vision gate ladder), built by a multi-agent Claude Code system with independent verifiers, git-hook gates and headless background lanes
Automator 6.5k tok · · Q49
samsalisbury/home
Home directory stuff, dotfiles and workflow tools.
Fort Knox 11 tok · · Q23
viberific/Autora
An agentic harness you can watch: the agent's browser, terminal, desktop and file edits stream into the conversation live, and every session records itself. Runs on Umbrel. Built with Claude Code.
Pragmatist 3.1k tok · · Q29
newell-paul/sprint-scout
Plan cycling segment attempts around the wind. A browser-only planner and a showcase of AI-native engineering.
Fort Knox 1.9k tok · · Q51