Getting started
ClawQueue is early. For v0.1.0, treat it as a trusted local tool for one operator running against GitHub repositories and projects they control.
Requirements
- Python 3.11+
- GitHub CLI (
gh) authenticated with access to your issue repo/project - A local runner you explicitly approve, such as OpenClaw, Claude Code, or Codex
- A GitHub issue repo and, optionally, a GitHub Project board
Install from source
bash
git clone https://github.com/ClawQueue/ClawQueue.git
cd ClawQueue
python3 -m pytest -qConfigure a private deployment
Copy the example config and fill in deployment-specific values outside public git history:
bash
cp config/clawqueue.private.example.json config/clawqueue.private.json
$EDITOR config/clawqueue.private.jsonPrivate values include repository names, ProjectV2 IDs, status option IDs, assignees, bot tokens, chat IDs, and local runner names.
Run one manual scheduler tick
bash
python3 scripts/status.py --no-queue
python3 scripts/scheduler.pyValidate that one eligible issue moves through the expected states before adding a periodic scheduler.
Next steps
- Learn the operator workflow
- Set up configuration
- Decide where artifacts should live