rabbyhub/rabby-mobile/mobile-pr-ready-watch
- community
Checking that the SKILL.md file is still in the repository.
Mark a Rabby Mobile GitHub PR ready for review, inspect thread-aware review state, watch for new review feedback, implement addressed fixes, and resolve review threads after fixes are pushed. Use when Codex is asked to make a mobile PR ready, keep observing a PR, handle GitHub review comments, or resolve addressed PR review threads.
Install
npx vskill@latest install rabbyhub/rabby-mobile/mobile-pr-ready-watchSKILL.md
Mobile PR Ready Watch
Use this skill for the repeatable "make the PR ready, then watch and handle review feedback" workflow.
Default watch window: 15 minutes. Poll every 2-3 minutes unless the user gives a different duration.
Rules
- Work from the Rabby Mobile repository checkout that owns the PR.
- Use the authenticated GitHub CLI or the repository's configured GitHub tooling.
- Do not print tokens, GitHub auth internals, or credential files.
- Do not use
git ls-remote, PR refs, or merge refs as a substitute for review inspection. Review comments require GitHub API access. - Do not merge the PR unless the user explicitly asks.
- Do not resolve a thread until the issue is fixed, confirmed obsolete, or explicitly accepted as not actionable.
Opening Checks
Identify the PR from the current branch unless the user gives a PR number or URL:
If the PR is draft and the request includes making it ready, mark it ready:
If GitHub auth is missing or invalid, stop and report the blocker. Do not start a timed watch without review-thread access.
Review Inspection
Fetch review threads through GraphQL so isResolved, isOutdated, line anchors, and threaded comments are visible:
Classify every unresolved thread:
needs-fix: the reviewer points out a real bug, regression, missing guard, missing test, or risky behavior.already-fixed: current code already addresses the comment.obsolete-outdated: the thread is outdated and the current code path no longer has the problem.needs-discussion: the comment conflicts with an explicit product decision or is too broad for the current iteration.
Surface needs-discussion items before editing. For needs-fix, implement the smallest focused fix.
Handling Feedback
For each actionable review thread:
- Read the referenced file and nearby code before editing.
- Patch only the reviewed issue and directly related tests.
- Preserve business semantics unless the user explicitly approves a behavior change.
- Run the narrowest meaningful verification first; broaden when the change touches shared behavior.
- Commit and push the fix before resolving the thread.
Resolve addressed review threads with GraphQL:
After resolving, re-fetch review threads and confirm the unresolved count.
Watch Loop
At each poll:
- Refresh PR metadata and review threads.
- Handle new unresolved feedback using the classification rules above.
- If code changes are made, verify, commit, push, then resolve only proven-addressed threads.
- Continue until the watch window ends or the user stops the watch.
When entering a long watch, give a short status note:
Final Report
Report:
- PR number and URL
- whether it was already ready or changed from draft
- review threads fixed and resolved
- unresolved threads left open and why
- commits pushed during the watch
- verification that passed or could not be run
Security
- Security scanPassed, 96 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/mobile-pr-ready-watch/SKILL.md
Popularity
- Repository stars
- 82
- Repository forks
- 37
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/rabbyhub/rabby-mobile/mobile-pr-ready-watch)