🛡️ Rigour

Local-first quality gates for AI-generated code.
Rigour forces AI agents to meet strict engineering standards before marking tasks "Done".

Zero cloud. Zero telemetry. PASS/FAIL is always free.

🚀 Quick Start

npx @rigour-labs/cli init     # Initialize quality gates
npx @rigour-labs/cli check    # Verify code quality
npx @rigour-labs/cli run -- claude "Build feature X"  # Agent loop

🛑 The "Vibe Coding" Trap

AI agents often fall into a cycle of "Guess and Hope":

1. Agent makes a change.
2. Agent claims "Task 100% complete" or "CI will pass now."
3. Agent pushes to remote.
4. CI Fails (Type error, lint failure, broken test).

This is "Vibe Coding"—the agent is hallucinating success based on narrative, not execution.

Rigour breaks this cycle. It forces the agent to face the same cold, hard verification tools (ruff, mypy, vitest) that CI runs, but locally and immediately. Rigour turns a "claim of victory" into a "proof of execution."

🔄 How It Works

Agent writes code → Rigour checks → FAIL? → Fix Packet → Agent retries → PASS ✓

The rigour run command loops until your agent achieves PASS or hits max iterations.

⚙️ What Gets Checked

🤖 Works With

• Claude Code: rigour run -- claude "..."
• Cursor / Cline / Gemini: Via high-fidelity MCP server (rigour_check, rigour_explain)

📖 Documentation

📚 Full Documentation →

🧪 CI Integration

- run: npx @rigour-labs/cli check --ci

See full example for GitHub Actions setup.

📜 License

MIT © Rigour Labs

"Rigour adds the engineering."