- Home
- Skills
- Agents & Orchestration
- Git PR Workflow
More screenshots
Works with the AI Agents you already use
Git PR Workflow
End-to-end git + PR automation: severity-based code review, conventional commits, secret scanning, and one-command ship to GitHub/GitLab.
$12
Git PR Workflow
Example session with this skill installed
I've finished the new user auth logic. Check for issues, then ship it to a new branch called feat/auth-flow.
- Read your context and instructions
- Compiled the git pr workflow
Pre-flight complete.
- No secrets detected in the diff
- Untracked
.env.localfound — will be ignored - Working tree is clean aside from the auth changes
- Base branch is
mainand up to date
Code Review Summary
Solid TypeScript implementation that follows existing project patterns. One medium-severity issue around refresh-token expiry handling should be fixed before merge.
| Severity | Location | Issue | Suggested Fix |
|---|---|---|---|
| Medium | src/auth/service.ts:42 | No explicit handling for expired refresh tokens | Catch TokenExpiredError and force re-login |
| Low | src/auth/types.ts:12 | Interface name is vague | Rename AuthData → UserSessionPayload |
| Positive | src/auth/hooks.ts | Clean use of memoized selectors for session state | — |
Proposed Commit
feat(auth): implement secure user authentication flow
- Add JWT service for session management
- Implement useSession hook for frontend state
- Add integration tests for login + refresh paths
Refs: #104
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Coding agents often leave dirty working trees, vague commit messages, and thin PR descriptions. Developers waste time manually staging files, hunting for leaked secrets, rewriting commit messages, and writing the same PR boilerplate over and over.
What it does
- Runs pre-flight safety checks (secrets, untracked files, merge conflicts, large binaries) before any git action.
- Produces structured, severity-based code reviews focused on correctness, security, and performance — not style nits.
- Generates strict Conventional Commit messages (and can split mixed diffs into atomic commits).
- Automates the full ship loop: branch → stage → commit → push → open PR (via
ghorglab). - Writes complete, reviewer-friendly PR descriptions using the repo’s own template or a clean default.
Frameworks & tools
Git, GitHub CLI (gh), GitLab CLI (glab), Bitbucket. Language-aware checks for Python, JavaScript/TypeScript, Go, and Rust.
Why this beats prompting it yourself
Generic prompts miss secrets, produce inconsistent commit styles, and generate weak PR descriptions. This skill enforces a senior-engineer workflow every time — so every change is audited, correctly categorized, and documented for humans.
Use cases
- “Ship it” — go from local changes to a remote PR in one flow.
- Audit a feature branch for security and logic issues before merge.
- Turn a messy diff into a clean sequence of conventional commits.
- Generate standardized PR descriptions that give reviewers real context.
Known limitations
Recommends human design review for diffs larger than ~1500 LOC or changes touching auth, payments, or migrations.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Recommends human design review for diffs larger than ~1500 lines of code, or for changes that touch critical infrastructure such as authentication, payments, or database migrations. Does not automatically run your test suite or linters (it will surface failures only if those commands are available in the environment).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago
- Software
Git