coding-agent
by @tshogx
Delegate coding tasks to Codex, Claude Code, or Pi agents via background process. Use when: (1) building/creating new features or apps, (2) reviewing PRs (sp...
mktemp -d && git init for scratch workexec is clean β codex exec "prompt" runs and exits, perfect for one-shotssubmit vs write β submit sends input + Enter; write sends raw data without newline1. Right execution mode per agent: Claude Code uses --print --permission-mode bypassPermissions (no PTY); Codex/Pi/OpenCode use pty:true
2. Respect tool choice β if the user asks for a specific agent, use it; don't silently substitute
3. Orchestrator mode β do not hand-code patches yourself when the task is delegated to an agent
4. Be patient β don't kill sessions just because they're slow; check with process:log first
5. Parallel is fine β run multiple agents at once for batch work
6. Never start agents in the OpenClaw workspace β they'll read soul docs and get confused about who's in charge
7. Never checkout branches in the live OpenClaw project folder β that's the running instance
clawhub install tshogx-coding-agent