claude-pty-agentslisted

# Claude PTY agents Use persistent Claude Code processes owned by this Codex thread. Launch as many as the work needs, including several in one canonical worktree; the launcher imposes no limit and never refuses a launch because another Claude process or registered worker shares the root. Keep Codex as the owner of intent, material architecture or product tradeoffs, authority, conflicts, independent verification, and the final verdict. Treat this skill as transport and custody policy, never as additional authority. ## Contract and boundaries Give an edit-capable worker one compact contract: - `Outcome`: state that must become true. - `Done when`: observable acceptance criteria. - `Boundaries`: exact root, side effects, prohibited actions, and ownership. - `Authoritative context`: applicable source-of-truth material and unknowns. - `Non-goals`: adjacent work not to absorb. - `Required handoff`: material evidence, risk, uncertainty, missing authority, deliberate non-actions, and custody needed for Codex to decide. The worker is permanently local-only. It cannot commit, push, publish, deploy, control services, send external messages, administer the host, operate on credentials, modify Claude/Codex configuration, or perform destructive remediation. Such work requires separate Codex review and exact current-user authorization. Maintain one edit owner per canonical worktree. The generated settings, deny rules, prompt, and hook are cooperative controls, not an OS sandbox. Ba