TurboKach/claude-code-setup
Claude Code setup: opinionated CLAUDE.md, a feature pipeline skill (plan → delegated execute → codex cross-review → ship), pinned role agents, parallel subagent fan-out, and a one-command installer with self-update.
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 TurboKach/claude-code-setup/agents ./rig-claude-code-setup/agents $ npx degit TurboKach/claude-code-setup/skills ./rig-claude-code-setup/skills $ npx degit TurboKach/claude-code-setup/hooks ./rig-claude-code-setup/hooks
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 TurboKach/claude-code-setup/skills/agent-teams .claude/skills/agent-teams $ npx degit TurboKach/claude-code-setup/skills/analyze-arcs .claude/skills/analyze-arcs $ npx degit TurboKach/claude-code-setup/skills/feature-workflow .claude/skills/feature-workflow $ npx degit TurboKach/claude-code-setup/skills/stack-update .claude/skills/stack-update
$ curl -fsSL --create-dirs -o .claude/agents/codex-triage.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/codex-triage.md $ curl -fsSL --create-dirs -o .claude/agents/explorer.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/explorer.md $ curl -fsSL --create-dirs -o .claude/agents/fixer.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/fixer.md $ curl -fsSL --create-dirs -o .claude/agents/plan-reviewer.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/plan-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/spec-reviewer.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/spec-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/step-executor.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/step-executor.md $ curl -fsSL --create-dirs -o .claude/agents/team-executor.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/team-executor.md $ curl -fsSL --create-dirs -o .claude/agents/team-merger.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/team-merger.md $ curl -fsSL --create-dirs -o .claude/agents/team-reviewer.md https://raw.githubusercontent.com/TurboKach/claude-code-setup/master/agents/team-reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(~/.aws/**)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(.env)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/.env.*)",
"Read(**/*.key)",
"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(npm publish:*)",
"Bash(wget *)",
"Bash(git rebase *)",
"Bash(gh pr merge *)",
"Bash(git commit *)"
]
}
} Skills (4)
Subagents (9)
| codex-triage model: sonnet | Feature-workflow review triage. Reads one round's `codex-challenge.sh` output file(s) — all slices of a split round go to the one spawn — verifies each finding against `git show <head>:<path>` and `gi |
| explorer model: sonnet | Read-only codebase search for the pipelines — the pinned stand-in for the built-in Explore agent (which floats with the session's effort and runs on Opus under a Fable or Opus master; this agent is pi |
| fixer model: sonnet | Feature-workflow fixer. Implements one review round's finding set — codex P0/P1 findings plus their adjacent P2s, playtest regressions — on the session's own branch, with no other writer running at th |
| plan-reviewer model: inherit | Plan validator for the feature-workflow and agent-teams pipelines. Reads the plan the lead wrote to the plan file and checks it against the actual codebase before the lead presents it via ExitPlanMode |
| spec-reviewer model: sonnet | Feature-workflow spec-conformance pass. At the final review gate, reads the approved plan file and the feature's whole diff range and reports gaps only — unimplemented or incomplete requirements, scop |
| step-executor model: sonnet | Feature-workflow executor. Implements one sequential step of an approved plan on the session's own branch, with no other writer running at the same time. Spawn UNNAMED (never pass name:) so its final |
| team-executor model: sonnet | Agent-teams executor. Implements one independent unit of an approved plan from a self-contained spawn prompt, running concurrently with sibling executors. Use only for parallel fan-out; a single seque |
| team-merger model: sonnet | Parallel-run merger. Merges the worktree branches the lead names into the base branch in turn — reviewer-approved units in an agent-teams run, a feature-workflow pair once both executors report — reso |
| team-reviewer model: opus | Agent-teams reviewer. Adversarially verifies each executed unit's diff before it merges — correctness, regressions, test gaps, scope creep. Use after executors finish and before the merger lands work. |
Similar rigs
theishandubey/claude-config
A complete Claude Code setup. An agent roster, an orchestration playbook, safe defaults and curated skills, installed with one command.
Orchestrator 4.0k tok ·
bi-boo/claude-code-config
我的 Claude Code 个人配置——CLAUDE.md 规则、Agent、记忆机制，以及 Ghostty 终端配置
YOLO Cowboy 1.4k tok ·
HGInsights/claude-code-setup
Production-tested Claude Code harness configuration: hooks, skills, subagents, and automated review loops
Pragmatist 1.9k tok ·
Dana-Afazeli/agentic-swe-kit
A maintained template for the agentic-SWE harness: proven gates, Claude Code hooks, CI, a headless review loop, and the workflow docs behind them. Use this template, run kit.py init, take updates with kit.py update.
Fort Knox 1.7k tok ·