Coordinate parallel coding agents and review their worktrees with Orca
Run Codex, Claude Code, OpenCode, Pi, and similar coding agents side by side in isolated worktrees, then compare, annotate, and merge the best result.
npx skills add agentskillexchange/skills --skill coordinate-parallel-coding-agents-and-review-their-worktrees-with-orca
Use Orca when a development workflow needs several coding agents running in parallel, each isolated in its own git worktree, with one operator supervising terminals, diffs, task context, and follow-up prompts. The upstream README frames Orca as an AI development environment for running Codex, Claude Code, OpenCode, or Pi side by side, with parallel worktrees, mobile monitoring, embedded browser design mode, GitHub and Linear task views, SSH worktrees, diff annotation, and an Orca CLI for scripted workflows.
What this skill actually does
Invoke this instead of using each agent product normally when the real job is orchestration: fan out a task, keep each agent isolated, compare outputs, annotate AI diffs, and merge the chosen result. The scope boundary is supervised multi-agent coding operations around worktrees and review. It is not a general desktop IDE listing, generic agent framework, or mobile companion app entry.
Inputs and prerequisites: Orca desktop app, Git worktrees, one or more supported coding agents such as Codex, Claude Code, OpenCode, or Pi, optional mobile companion, optional SSH host.
Setup notes: Download Orca from https://onorca.dev/download, connect the coding agents you use, then create isolated worktrees for parallel agent runs. Use the Orca CLI for scripted worktree, snapshot, click, and fill workflows.
Source and verification boundary: use https://onorca.dev as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.