worktreeslisted

<!-- requires: none --> <!-- chains: none --> <!-- generated from .ai/workflows/worktrees.md; do not edit directly --> > **Codex:** this skill is the `/worktrees` adapter. Use Codex role agents from `.codex/agents/` when the runtime exposes them; otherwise perform the work directly and keep the same structured output. > Codex does not substitute `$ARGUMENTS` — read it as the arguments given in the user's request. Orchestrate parallel multi-session development: survey existing worktrees, reap finished ones, plan a wave of non-conflicting issues, create sibling worktrees, and spawn a seeded Codex session per issue. Run autonomously. Modes from `$ARGUMENTS`: no args = full cycle (survey → reap → plan → create → spawn). `status` = survey report only. `cleanup` = survey + reap only. ## Step 1 — Config Detect the repo (`git remote get-url origin`) and read the optional `## Worktrees` section of the project's CLAUDE.md: - `max:` — concurrent-session cap (default **3**) - `serialize_paths:` — files where issues collide **semantically** (a shared schema, an installer, a generated bundle); issues likely touching the same listed path are never scheduled in the same wave. Do not list a file every PR touches by construction (a changelog, a version manifest) — serializing it blocks every pair and pins concurrency at 1. Fix those structurally instead, with one file per PR. - `setup:` — per-worktree setup command (deps, env), run after creation ## Step 2 — Survey `git worktree list -