Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
94 lines (68 loc) · 3.34 KB

CONTRIBUTING.md

File metadata and controls

94 lines (68 loc) · 3.34 KB
Raw

Contributing to Atelier Pipeline

Thank you for your interest in contributing to Atelier Pipeline, a multi-agent orchestration system for AI-powered IDEs. This project is licensed under Apache 2.0.

Code of Conduct

All participants are expected to follow our Code of Conduct. Be respectful, constructive, and inclusive.

Reporting Bugs

Found a bug? Please open a bug report.

Include:

  • A clear description of the problem
  • Steps to reproduce
  • Expected vs. actual behavior
  • Your environment (OS, shell, Node.js version, IDE)

Suggesting Features

Have an idea? Open a feature request.

Describe the use case and the problem it solves. Concrete examples help.

Security Vulnerabilities

Do not open a public issue for security vulnerabilities. See SECURITY.md for responsible disclosure instructions.

Development Setup

  1. Clone the repository:

    git clone https://github.com/robertsfeir/atelier-pipeline.git
cd atelier-pipeline

  2. Run the /pipeline-setup skill inside Claude Code to install pipeline files into your project's .claude/ directory.

  3. Run the test suite:

    bats tests/hooks/ && cd brain && node --test ../tests/brain/*.test.mjs

    Hook tests use Bats. Brain tests use the Node.js built-in test runner.

Architecture Overview

Atelier Pipeline is a multi-agent system where Eva (the Pipeline Orchestrator) routes work across 11+ specialized agents:

Agent Role
Robert Product -- feature discovery, specs, product strategy
Sable UX -- user experience and interaction design
Cal Architecture -- ADR production and technical planning
Colby Build -- implementation and coding
Roz QA -- test authoring, validation, and investigation
Ellis Commits -- git operations and changelog
Agatha Documentation -- doc planning and writing
Poirot Review -- blind diff-based code review
Sentinel Security -- Semgrep-backed SAST auditing
Deps Dependencies -- outdated scan, CVE checks, upgrade risk
Darwin Evolution -- telemetry analysis and pipeline improvement

Source templates live in source/ and are installed to .claude/ (Claude Code) or .cursor-plugin/ (Cursor) by the setup skill. The brain/ directory contains the Atelier Brain MCP server (Node.js + PostgreSQL).

Pull Request Guidelines

  • Branch from main. Create a feature or fix branch off of main.
  • Tests must pass. Run the full test suite before submitting.
  • Follow existing patterns. Match the conventions in CLAUDE.md and the surrounding code. Roz-first TDD, Eva never writes code, ADRs are immutable.
  • One concern per PR. Keep pull requests focused on a single change. Split unrelated work into separate PRs.
  • Link related issues. Reference the issue number in your PR description (e.g., "Fixes #42").
  • Describe what and why. The PR description should explain the change and its motivation, not just list the files touched.

Maintainer

This project is maintained by Robert Sfeir.