{"id":50142277,"title":"Show HN: Pi-cowboy – use Codex, Claude Code, OpenCode as subagents","points":3,"user":"suoloordi","time":1792339497,"time_ago":"9 hours ago","type":"link","content":"

pi-cowboy is an agent orchestrator for Pi(https://pi.dev) and Herdr(https://github.com/herdrdev/herdr). Break large jobs into focused subtasks, hand them off to background agents running in parallel, and consolidate the results when they finish. Each agent runs in an isolated copy-on-write Git work-tree, minimising disk overhead.

pi-cowboy can orchestrate foreign harnesses too, including Claude Code, Codex, OpenCode and more.

Feature highlights:

- Foreign harnesses: \n pi-cowboy agents can run in Claude Code, Codex, OpenCode, Pig and Pi-Bolt, as well as Pi. This gives you great flexibility in choosing how to spend your tokens. Running out of Codex usage? Switch your pi-cowboy harness to OpenCode, select one of its free models, and let it handle the rest.\n Pig and Pi-Bolt are pi variants that are more efficient, with a lower memory footprint and faster performance than pi. Because pi-cowboy is designed to run multiple agents in parallel, having a lower per-agent footprint is crucial. It prevents your machine from running out of resources and lets you have higher concurrency without choking your system.

- Concurrency limits: \n You can configure concurrency limits at every level. Running a local model with limited resources? If a simple task already takes hours to complete, the last thing you want is to overload your machine by running too many of them at once. Set concurrency limits for that model specifically to keep resource usage under control.

You can install it on Pi as a package:

  pi install npm:pi-cowboy #(requires `herdr` 0.8+)

","url":"https://github.com/electric-dragon-80000v/pi-cowboy","domain":"github.com","comments":[],"comments_count":0}