babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep watching open PRs so fresh review feedback is surfaced promptly. Use when the user asks Codex to monitor a PR, watch CI, handle review comments, or keep an eye on failures and feedback on an open PR.
79
80%
Does it follow best practices?
Impact
95%
1.72xAverage score across 2 eval scenarios
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./.codex/skills/babysit-pr/SKILL.mdEvaluation results
100%
62%PR Action Plan
PR babysitting action plan from watcher snapshot
Review before CI retry
100%
100%
Review fix commit message
0%
100%
CI fix commit message format
0%
100%
No auto-reply to human comment
13%
100%
User-approved reply prefix
0%
100%
Thread resolution comment prefix
0%
100%
Restart watch after push
60%
100%
Continue monitoring when green
100%
100%
Reads actions field
100%
100%
Thread resolution scope
100%
100%
90%
18%PR Babysitter Automation Script
PR monitoring loop script with correct watcher flags, stop conditions, and commit messages
Correct watcher path
100%
90%
--watch as primary mode
100%
100%
--pr auto flag
100%
100%
No concurrent watchers
100%
75%
Restart after push
100%
87%
Terminal ownership
100%
100%
Autonomous polling
100%
100%
Correct stop conditions
62%
100%
No stop on idle
66%
66%
No stop on green
0%
100%
Polling cadence
42%
42%
CI fix commit message
14%
100%
Review fix commit message
12%
100%
- Repository
- openai/codex
- Commit
78245b4
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.