Test Generation at Scale

One agent per module. Full test coverage by morning.

Low test coverage? Assign one Claude Code agent per module or file. Each agent writes unit tests, integration tests, and edge cases. amux's self-healing ensures they keep working through context resets.

Workflow

1. List your untested modules
2. Post each as a task on the amux board
3. Start agents: they claim tasks atomically from the board
4. Monitor from the dashboard — check coverage as agents complete work

Why self-healing matters here

Test generation is token-intensive. Agents frequently hit context limits. amux's watchdog automatically runs /compact and restarts agents that crash — so your test generation runs don't stall at 3am.

git clone https://github.com/mixpeek/amux && cd amux && ./install.sh
amux register myproject --dir ~/Dev/myproject --yolo
amux start myproject
amux serve  # → https://localhost:8822