pull-request
Codex compatibility note:
- Invoke
$skill-name: load the registered.agents/skills/<name>/SKILL.mdand execute it with Codex tools; no separateSkilltool or Claude mode switch is needed. Author.claude/**; mirrors are generated.- Track steps before execution and keep progress/evidence current. Explicit skill requests execute their protocol. Use the native question tool for required questions.
- Map foreign tool names to available host capabilities. A name mismatch needs no approval; an actually unavailable required capability stops with its step/evidence and a user question before adaptation.
- Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required
spawn_agentsubagent(s) for that task.- Preserve protocol order unless the user approves a deviation. Workflows use
$start-workflow: gates stay fixed; other steps may flex with logged reasons and evidence.
Requested Git operations
An explicit request to commit or push authorizes that operation once its required checks pass. Do not ask for a second commit/push confirmation, offer an unrequested commit/push, or end with an offer to perform the already requested operation. A commit request does not include push. Run required test and review decisions automatically; a failed gate reports the concrete blocker rather than asking to waive it. Scope ambiguities and destructive actions outside the request still need resolution. Native host permission prompts are separate from this skill.
Quick Summary
Goal: Drive authorized work to a ready-to-merge PR with a whole-branch review and current local test evidence, both chosen automatically by risk, and green CI; never ask the user whether to run tests or review, and stop with a report on a real blocker.
Step 3.3 keeps the project's tracked work current with the pull request, as the project's own work-tracking instruction says; it never gates the PR.
Summary: