auerbachb/claude-code-config
CLAUDE.md config that teaches Claude Code to collaborate with CodeRabbit for automated PR planning, code review loops, and merge workflows
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit auerbachb/claude-code-config/.claude ./rig-claude-code-config # 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.
$ npx degit auerbachb/claude-code-config/.claude/skills/admin-merge .claude/skills/admin-merge $ npx degit auerbachb/claude-code-config/.claude/skills/babysit-pr-stop .claude/skills/babysit-pr-stop $ npx degit auerbachb/claude-code-config/.claude/skills/babysit-pr .claude/skills/babysit-pr $ npx degit auerbachb/claude-code-config/.claude/skills/board .claude/skills/board $ npx degit auerbachb/claude-code-config/.claude/skills/end-resume .claude/skills/end-resume $ npx degit auerbachb/claude-code-config/.claude/skills/end .claude/skills/end $ npx degit auerbachb/claude-code-config/.claude/skills/fixpr .claude/skills/fixpr $ npx degit auerbachb/claude-code-config/.claude/skills/go-on .claude/skills/go-on $ npx degit auerbachb/claude-code-config/.claude/skills/harness-audit .claude/skills/harness-audit $ npx degit auerbachb/claude-code-config/.claude/skills/issue-maker .claude/skills/issue-maker $ npx degit auerbachb/claude-code-config/.claude/skills/leave-by .claude/skills/leave-by $ npx degit auerbachb/claude-code-config/.claude/skills/memory-clean .claude/skills/memory-clean $ npx degit auerbachb/claude-code-config/.claude/skills/merge-conflict .claude/skills/merge-conflict $ npx degit auerbachb/claude-code-config/.claude/skills/merge .claude/skills/merge $ npx degit auerbachb/claude-code-config/.claude/skills/pause-resume .claude/skills/pause-resume $ npx degit auerbachb/claude-code-config/.claude/skills/pause .claude/skills/pause $ npx degit auerbachb/claude-code-config/.claude/skills/pm-clean .claude/skills/pm-clean $ npx degit auerbachb/claude-code-config/.claude/skills/pm-forgotten-pr .claude/skills/pm-forgotten-pr $ npx degit auerbachb/claude-code-config/.claude/skills/pm-handoff .claude/skills/pm-handoff $ npx degit auerbachb/claude-code-config/.claude/skills/pm .claude/skills/pm $ npx degit auerbachb/claude-code-config/.claude/skills/pr-monitor-and-manage-stop .claude/skills/pr-monitor-and-manage-stop $ npx degit auerbachb/claude-code-config/.claude/skills/pr-monitor-and-manage-wake .claude/skills/pr-monitor-and-manage-wake $ npx degit auerbachb/claude-code-config/.claude/skills/pr-monitor-and-manage .claude/skills/pr-monitor-and-manage $ npx degit auerbachb/claude-code-config/.claude/skills/prompt .claude/skills/prompt $ npx degit auerbachb/claude-code-config/.claude/skills/quotas-setup .claude/skills/quotas-setup $ npx degit auerbachb/claude-code-config/.claude/skills/quotas .claude/skills/quotas $ npx degit auerbachb/claude-code-config/.claude/skills/recap .claude/skills/recap $ npx degit auerbachb/claude-code-config/.claude/skills/receiving-code-review .claude/skills/receiving-code-review $ npx degit auerbachb/claude-code-config/.claude/skills/review-stack-audit .claude/skills/review-stack-audit $ npx degit auerbachb/claude-code-config/.claude/skills/standup .claude/skills/standup $ npx degit auerbachb/claude-code-config/.claude/skills/start-issue .claude/skills/start-issue $ npx degit auerbachb/claude-code-config/.claude/skills/status .claude/skills/status $ npx degit auerbachb/claude-code-config/.claude/skills/subagent .claude/skills/subagent $ npx degit auerbachb/claude-code-config/.claude/skills/wave .claude/skills/wave $ npx degit auerbachb/claude-code-config/.claude/skills/wrap .claude/skills/wrap
$ curl -fsSL --create-dirs -o .claude/agents/phase-a-fixer.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/phase-a-fixer.md $ curl -fsSL --create-dirs -o .claude/agents/phase-b-reviewer.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/phase-b-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/phase-c-merger.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/phase-c-merger.md $ curl -fsSL --create-dirs -o .claude/agents/pm-worker.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/pm-worker.md $ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/researcher.md $ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/auerbachb/claude-code-config/main/.claude/agents/silent-failure-hunter.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (35)
admin-mergebabysit-pr-stopbabysit-prboardend-resumeendfixprgo-onharness-auditissue-makerleave-bymemory-cleanmerge-conflictmergepause-resumepausepm-cleanpm-forgotten-prpm-handoffpmpr-monitor-and-manage-stoppr-monitor-and-manage-wakepr-monitor-and-managepromptquotas-setupquotasrecapreceiving-code-reviewreview-stack-auditstandupstart-issuestatussubagentwavewrap
Subagents (6)
| phase-a-fixer model: opus | Phase A subagent: fix review findings, push code, write handoff file, print exit report. Used after a PR receives CR/BugBot/Greptile review findings. |
| phase-b-reviewer model: opus | Phase B subagent: poll for CR/BugBot/Greptile reviews, process findings, fix code, update handoff file, print exit report. Runs after Phase A pushes fixes. |
| phase-c-merger model: sonnet | Phase C subagent: verify merge gate and AC, then run the canonical /wrap merge flow when authorized. |
| pm-worker model: sonnet | PM task execution agent: issue management and repo bootstrap checks. Used for lightweight PM tasks that don't require the full Phase A/B/C pipeline. |
| researcher model: sonnet | Read-only research subagent: explore, audit, or investigate the codebase and GitHub state without any risk of file modification. Use when you need findings, not fixes. |
| silent-failure-hunter model: sonnet | Read-only review agent that hunts silent failures in Bash scripts and shell tooling — swallowed exit codes, fabricated sentinels, guarded no-ops, and missing error propagation. Adapted from affaan-m/e |
Similar rigs
ozzaii/codex-claude-loop
Gated Claude x Codex build loop for Claude Code: Codex plans and implements from one persistent thread, Claude approves the plan and reviews the diff. Zero deps, 20-test smoke suite.
Pragmatist 216 tok ·
AndreAlbuquerque/claude-config
Personal Claude Code config — global CLAUDE.md, agents (@pm, @visionary, @discovery, @designer, @engineer, @implementer, @researcher, @reviewer, @qa), and skills, synced across machines.
Orchestrator 3.0k tok ·
karanb192/claude-code-hooks
🪝 Claude Code hooks + an installable plugin marketplace: safety, cost, observability, productivity.
Pragmatist 268 tok ·
onurpolat05/ALBA
Turn Claude Code into a personal AI agent with persistent memory. Interactive setup, 9 skills, 8 hooks, role-agnostic. Verified against v2.1.220.
Fort Knox 3.9k tok ·