iadr-dev/greenlight
AI × BDD spec-driven development for Claude Code. greenlight turns a feature request into human-approved, executable specifications, and then into working code. People approve the contract (rules and examples); the AI does the mechanical wo
stack: Python
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 iadr-dev/greenlight/agents ./rig-greenlight/agents $ npx degit iadr-dev/greenlight/skills ./rig-greenlight/skills $ npx degit iadr-dev/greenlight/hooks ./rig-greenlight/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 iadr-dev/greenlight/skills/adopt .claude/skills/adopt $ npx degit iadr-dev/greenlight/skills/approve .claude/skills/approve $ npx degit iadr-dev/greenlight/skills/change .claude/skills/change $ npx degit iadr-dev/greenlight/skills/feature .claude/skills/feature $ npx degit iadr-dev/greenlight/skills/init .claude/skills/init $ npx degit iadr-dev/greenlight/skills/status .claude/skills/status
$ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/product-analyst.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/product-analyst.md $ curl -fsSL --create-dirs -o .claude/agents/qa-designer.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/qa-designer.md $ curl -fsSL --create-dirs -o .claude/agents/spec-auditor.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/spec-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/step-engineer.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/step-engineer.md $ curl -fsSL --create-dirs -o .claude/agents/system-analyst.md https://raw.githubusercontent.com/iadr-dev/greenlight/main/agents/system-analyst.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 (6)
Subagents (6)
| implementer model: sonnet | Writes the production code that turns the red scenarios green, one task at a time (implement phase). Writes production code only; the features, step definitions and drivers are read-only. |
| product-analyst model: opus | Turns a feature request into a brief and approved-ready rules (discover and specify phases). Use when greenlight starts a feature or when rules must be written or revised. Writes only under .bdd/. |
| qa-designer model: opus | Designs the tests of a feature from its approved rules, then writes the Gherkin examples (test-design and examples phases). Use when examples must be derived, written or revised. Writes under .bdd/ an |
| spec-auditor model: opus | Independent review of a finished feature against the greenlight checklist, from artifacts only, returning a report with evidence for every line (verify phase). Read-only: it writes nothing and returns |
| step-engineer model: sonnet | Resolves the step vocabulary and binds the scenarios to step definitions and drivers, reaching a valid red before any production code exists (refine and bind phases). Writes under .bdd/ and the test, |
| system-analyst model: opus | Maps every step of the approved scenarios to the interface it exercises in the code base (system-map phase). Use when a feature's examples are approved and the system map must be written. Reads the wh |
Similar rigs
kirthar/sdd-kit
Spec-driven development for Claude Code: size-based routes, human approval gates, hook-enforced guardrails and measured with evals.
Skill Collector 1.8k tok · · Q5
KTCrisis/flux7-mesh
Guardrail sidecar proxy for AI agents — policy, approval gates, grants, tracing. One Go binary, one YAML.
Pragmatist 126 tok · · Q0
front-factory/brigade
Claude Code, organized like a kitchen brigade.
Fort Knox 400 tok · · Q26
vamsivarma27/proofloop
Your coding agent says done. ProofLoop makes it prove it. Evidence-gated engineering harness and Claude Code plugin.
Orchestrator 1.4k tok · · Q3