AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Work With PR — Full PR Lifecycle

Splits a task into atomic PRs, builds each in its own git worktree, and loops through CI and bot review gates until merged.

Dev & CodingAdvanced★ 67,907⑂ 5,550AI score 8/10Last updated: Aug 15, 2026

What it does

  • Decomposes one task into the smallest independently-mergeable PRs, ordering independent slices for parallel work and dependent slices as a stack.
  • Creates a dedicated git worktree per PR as a sibling of the repo, so your main working directory (and any uncommitted work) is never touched.
  • Delegates implementation to the ulw-loop skill and enforces evidence-bound manual QA written to disk for every success criterion — no evidence file, no commit or push.
  • Generates a reviewer-friendly English PR body: Summary / Changes / QA & Evidence / Risks & Residuals / Related Issues, with guidance on sanitizing logs.
  • Runs an unbounded verification loop over two gates — CI (gh pr checks) and the Cubic review bot (skippable only on exhausted quota) — routing failures back to fix-and-re-QA.
  • Arms auto-merge (merge commit only), syncs .omo state back, then removes the worktree and reports iterations and gate outcomes.

Who it's for

  • Teams shipping through GitHub PRs with the gh CLI and a real review gate.
  • Developers who want an agent forced into small, atomic PRs instead of one 2000-line dump.
  • Repos with a similar pipeline: dev base branch, bun tooling, automated review bot.

Examples

  • "Implement this issue and open a PR" → branch + worktree → implement with QA evidence → PR → CI/Cubic green → auto-merge → worktree cleanup.
  • "Split this refactor into atomic PRs and run them in parallel" → one background subagent and worktree per independent PR.
  • CI fails on typecheck → pull gh run view --log-failed, fix only that issue, atomic commit, push, re-enter from Gate A.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/HEAD/.agents/skills/work-with-pr/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the .agents/skills/work-with-pr folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/work-with-pr/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent /tmp/omo && mkdir -p ~/.claude/skills && cp -r /tmp/omo/.agents/skills/work-with-pr ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Clone the repository: git clone https://github.com/code-yeongyu/oh-my-openagent /tmp/omo
  2. Make the skills folder: mkdir -p ~/.claude/skills
  3. Copy the skill: cp -r /tmp/omo/.agents/skills/work-with-pr ~/.claude/skills/
  4. Verify prerequisites: gh auth status (authenticated GitHub CLI), git --version, and bun --version if your project uses bun.
  5. Edit the repo-specific bits to match your project: BASE_BRANCH="dev", the bun typecheck/test/build commands, the Cubic bot gate, and the .omo sync step. Note it also expects a companion ulw-loop skill.
  6. Restart Claude Code, then trigger it with a request like "implement this and open a PR".
View source on GitHub ↗License: NOASSERTION