Welcome contributors — start here

Welcome to CoCo. This issue is your starting point.

What is CoCo?

An open-core AI orchestration framework: skills, agents, slash commands, and multi-agent orchestration, vendor-neutral across Claude Code, Cursor, Codex, and any AGENTS.md tool. The core is MIT; the Super Intelligence system (systems/superintelligence/) is proprietary and source-available.

How to contribute

Quick orientation

git clone https://github.com/coco-research/coco.git
cd coco
bash install.sh

Project state (v1.2.0)

• 6 system bundles: GSD, Brain, Team, Super Intelligence, HyperFrames, Cognee — each opt-in via --systems <name>
• 389 expert personas across 9 departments, powering the /SI-* advisory councils
• 242 Super Intelligence slash commands generated locally at install time from the team registries
• 4 IDE adapters (claude-code, cursor, codex, generic); CI runs a frontmatter linter, index verification, and command-reference checks on every push

Exact asset counts change with every merged skill, so rather than restate them here, the generated indexes are the source of truth:
skills/INDEX.md · commands/INDEX.md · agents/INDEX.md. They are regenerated by python3 scripts/build-index.py, which CI enforces.

Currently looking for help with

See the pinned good first issues. They range from documentation to small skill additions. Adapter work is a good entry point: see #4 (VS Code / Continue), #25 (Antigravity), #28 (Zed), and #26 (coco doctor).

Code of Conduct

By participating you agree to the Code of Conduct.

This issue is an orientation page rather than a defined change request. Read CONTRIBUTING.md and docs/architecture.md first, then inspect install.sh and scripts/build-index.py if the contributor guidance needs updating. No specific completion condition or target file is provided beyond this welcome content.