~ / rigs / pierrebelin / claude-code-toolkit

pierrebelin/claude-code-toolkit

Stop babysitting your coding agent. A Claude Code plugin that turns a business need into a spec, a plan and test-first code, audited before you commit. Built for DDD / clean-architecture repos.

↗ GitHub ★ 7 MIT updated 8d ago project Claude Code Claude plugin
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~3.7k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
1/5
No YOLO mode · details
Quality 20/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit pierrebelin/claude-code-toolkit/.claude ./rig-claude-code-toolkit  # 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 pierrebelin/claude-code-toolkit/presets/clean-architecture/skills/tests-contract-tests .claude/skills/tests-contract-tests
$ npx degit pierrebelin/claude-code-toolkit/presets/clean-architecture/skills/tests-e2e-tests .claude/skills/tests-e2e-tests
$ npx degit pierrebelin/claude-code-toolkit/presets/clean-architecture/skills/tests-integration-tests .claude/skills/tests-integration-tests
$ npx degit pierrebelin/claude-code-toolkit/presets/clean-architecture/skills/tests-unit-tests .claude/skills/tests-unit-tests
$ npx degit pierrebelin/claude-code-toolkit/skills/ask .claude/skills/ask
$ npx degit pierrebelin/claude-code-toolkit/skills/business-spec .claude/skills/business-spec
$ npx degit pierrebelin/claude-code-toolkit/skills/fix-bug .claude/skills/fix-bug
$ npx degit pierrebelin/claude-code-toolkit/skills/implement-tdd .claude/skills/implement-tdd
$ npx degit pierrebelin/claude-code-toolkit/skills/kit-init .claude/skills/kit-init
$ npx degit pierrebelin/claude-code-toolkit/skills/knowledge-merge .claude/skills/knowledge-merge
$ npx degit pierrebelin/claude-code-toolkit/skills/knowledge-sync .claude/skills/knowledge-sync
$ npx degit pierrebelin/claude-code-toolkit/skills/learn .claude/skills/learn
$ npx degit pierrebelin/claude-code-toolkit/skills/plan-implementation .claude/skills/plan-implementation
$ npx degit pierrebelin/claude-code-toolkit/skills/quality-report .claude/skills/quality-report
$ npx degit pierrebelin/claude-code-toolkit/skills/verify-ddd-tdd .claude/skills/verify-ddd-tdd
$ curl -fsSL --create-dirs -o .claude/agents/adversarial-reviewer.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/adversarial-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/ddd-tdd-auditor.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/ddd-tdd-auditor.md
$ curl -fsSL --create-dirs -o .claude/agents/knowledge-extractor.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/knowledge-extractor.md
$ curl -fsSL --create-dirs -o .claude/agents/knowledge-merger.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/knowledge-merger.md
$ curl -fsSL --create-dirs -o .claude/agents/tdd-cycle.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/tdd-cycle.md
$ curl -fsSL --create-dirs -o .claude/agents/tdd-implementer.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/tdd-implementer.md
$ curl -fsSL --create-dirs -o .claude/agents/tdd-test-author.md https://raw.githubusercontent.com/pierrebelin/claude-code-toolkit/main/agents/tdd-test-author.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 *)"
    ]
  }
}

Skills (15)

Subagents (7)

adversarial-reviewer
model: opus
Tries to refute a spec, a plan or a batch sheet before code is written, when /business-spec or /plan-implementation delegates it. Reading only; returns closed questions ranked Blocking or Major.
ddd-tdd-auditor
model: sonnet
Audits a .NET/DDD batch without ever modifying it when /verify-ddd-tdd is invoked. Reading, searching and running validations only.
knowledge-extractor
model: sonnet
Drafts one bounded context of the knowledge base from the code when /knowledge-sync delegates it — staged drafts with provisional ids, never the base itself.
knowledge-merger
model: sonnet
Drafts the merge of a delivered feature into the knowledge base when /knowledge-merge or the run-plan workflow delegates it — the spec's delta rewritten in the base's format, in the merge staging, nev
tdd-cycle
model: sonnet
Writes the RED tests then the GREEN code of one TDD cycle in a single agent when the run-lot workflow runs a light batch; a checkpoint checked on the tree separates the two phases.
tdd-implementer
model: sonnet
Writes the minimal production code that turns a red test green when /implement-tdd delegates the GREEN phase, then cleans up and states the access cost.
tdd-test-author
model: sonnet
Writes the RED tests for a .NET/DDD behaviour when /implement-tdd delegates the test-first phase.

Plugins (3)

context-band@cctoolkitknowledge-pane@cctoolkitrun-lot-pane@cctoolkit

Similar rigs

copied ✓