komoroko/loose-rein-kit
Human-on-the-Loop development harness: frozen requirement claims, blind actual-behaviour review, and human-confirmed gate approval as an installable CLI
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
Copy this rig
# review before running: this installs third-party code
$ npx degit komoroko/loose-rein-kit/.claude ./rig-loose-rein-kit # 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 komoroko/loose-rein-kit/.agents/skills/design .claude/skills/design $ npx degit komoroko/loose-rein-kit/.agents/skills/onboard .claude/skills/onboard $ npx degit komoroko/loose-rein-kit/.agents/skills/req .claude/skills/req $ npx degit komoroko/loose-rein-kit/.agents/skills/revise .claude/skills/revise $ npx degit komoroko/loose-rein-kit/.agents/skills/status .claude/skills/status $ npx degit komoroko/loose-rein-kit/.agents/skills/tasks .claude/skills/tasks $ npx degit komoroko/loose-rein-kit/.agents/skills/verify .claude/skills/verify $ npx degit komoroko/loose-rein-kit/.gemini/skills/adversarial-reviewer .claude/skills/adversarial-reviewer $ npx degit komoroko/loose-rein-kit/.gemini/skills/architect .claude/skills/architect $ npx degit komoroko/loose-rein-kit/.gemini/skills/implementer .claude/skills/implementer $ npx degit komoroko/loose-rein-kit/.gemini/skills/requirements-analyst .claude/skills/requirements-analyst
$ curl -fsSL --create-dirs -o .claude/agents/adversarial-reviewer.md https://raw.githubusercontent.com/komoroko/loose-rein-kit/main/.claude/agents/adversarial-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/komoroko/loose-rein-kit/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/komoroko/loose-rein-kit/main/.claude/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/requirements-analyst.md https://raw.githubusercontent.com/komoroko/loose-rein-kit/main/.claude/agents/requirements-analyst.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "Write|Edit|MultiEdit|NotebookEdit",
"hooks": [
{
"type": "command",
"command": "rein guard"
}
]
}
]
}
} Skills (11)
Subagents (4)
| adversarial-reviewer | Independent red-team critique of the mandate's material before it is approved. Attacks for counterexamples, ambiguities, hidden assumptions, and plan defects; report-only. Delegated from /req, /design |
| architect | In charge of the design phase. Builds an implementation approach from approved requirements and presents important technical choices as "options with trade-offs". Delegated from /design. |
| implementer | Implements a single task ticket (T-NNN), writes unit/integration tests that satisfy its acceptance criteria, and gets them green. Delegated one task at a time from the /build loop. |
| requirements-analyst | A sounding board for the requirements phase. Structures "what we want to achieve" from the product vision and points out gaps, ambiguities, and contradictions. Delegated from /req. |
Hooks (2)
| event | matcher | runs |
|---|---|---|
| SessionStart | * | test -f .rein/state.yaml && echo '--- Loose Rein: what changed since you last looked ---' && rein start || true |
| PreToolUse | Write|Edit|MultiEdit|NotebookEdit | rein guard |
Slash commands (8)
/build/design/onboard/req/revise/status/tasks/verify
Permissions
deny (0)
—
ask (0)
—
allow (30)
Read
Grep
Glob
Bash(cat .rein/config.yaml:*)
Bash(git status:*)
Bash(git diff:*)
Bash(git log:*)
Bash(git show:*)
Bash(git branch:*)
Bash(git rev-parse:*)
Bash(git add:*)
Bash(git commit:*)
Bash(git merge --ff-only:*)
Bash(git merge --no-ff:*)
Bash(git worktree:*)
Bash(ls:*)
Bash(rein start:*)
Bash(rein next:*)
Bash(rein dag:*)
Bash(rein build:*)
Bash(rein doctor:*)
Bash(rein events:*)
Bash(rein revise:*)
Bash(rein issue-sync:*)
Bash(rein sync --check:*)
Bash(make test:*)
Bash(make check:*)
Bash(python -m pytest:*)
Bash(make audit:*)
Bash(uv run pytest:*)
Similar rigs
Trolzie/claude-loop-systems-demo
A small demo of Markdown-as-system AI dev loops with Claude Code: issues → agents → skills → verification.
Pragmatist 480 tok · · Q15
Dcastroro/sdd-starter
Local-first Spec-Driven Development template and Claude Code plugin: traceable requirements, scoped implementation workflows and deterministic validation.
Pragmatist 591 tok · · Q3
sohaibdevv/Claude-Starter-Kit
A pre-built configuration that gives Claude Code persistent memory, security guardrails, and structured workflows. Clone, run setup.sh, and your next session picks up where the last one left off.
Pragmatist 435 tok · · Q0
procoders/superpowers-v
Compound V — a multi-model orchestrator for Claude Code, layered on Superpowers: a deterministic router (Opus for risky work, Sonnet for the mechanical), cross-model review (Claude + Codex), 3 scouts (code · docs · domain), semantic memory,
Pragmatist 3.0k tok · · Q15