PR Description Pack — Context + Test Plan
Turn commits or a diff into a reviewer-ready PR: summary, why, what changed, test plan, risks, and refs. Evidence-first — no invented tickets or fake CI greens. Cygnix Labs.
$5
Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsPR Description Pack — Context + Test Plan
Example session with this skill installed
Draft a PR from these commits: 'feat: add rate limiting to /api/search', 'fix: resolve memory leak in cache layer'. Intent is to stabilize search under high load. Issue #442.
- Read your context and instructions
- Compiled the pr description
I have analyzed your commits and intent. Here is a decision-useful PR draft based on the provided evidence.
Pull request draft
Recommended title
Implement search rate limiting and fix cache memory leaks
Description
Summary
Stabilizes the search endpoint by introducing request throttling and resolving a persistent memory leak in the internal cache layer.
Why
Prevents service degradation during high-traffic periods and ensures long-term process stability by stopping heap growth in the caching service.
What changed
- Rate Limiting: Added middleware to
api/searchto enforce request quotas. - Memory Fix: Refactored the cache layer to properly evict expired entries, closing the observed leak.
- Configuration: New environment variables for search rate thresholds.
Test plan
- Suggested: Load test
/api/searchwith concurrent requests to verify 429 responses. - Suggested: Monitor heap usage over 1 hour under synthetic load to confirm leak resolution.
Risks / rollout
- Risk: Legitimate high-volume users might be throttled.
- Rollout: Monitor search error rates immediately after deployment.
Issue / ticket refs
- #442
Why this shape
- Evidence used: Commit subjects and user intent.
- Uncertainties: Specific rate limits were not in the commit messages; inferred from typical API patterns.
Title alternatives
- Stabilize search API: Rate limiting and cache leak fix
- Fix search memory leak and add request throttling
- Search API performance and stability improvements
Red flags / ask author
- Confirm the specific rate limit values (TPS/RPM) intended for the search endpoint.
Next steps
- Select your preferred title from the alternatives above.
- Verify the suggested test plan against your local environment.
- Paste the Markdown into your VCS provider.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
Writing PR descriptions is a chore that leads to lazy summaries like "misc fixes" or "update files," which slows down reviews and hides risks. This skill transforms raw git evidence into decision-useful documentation for busy maintainers. It eliminates the friction of manual drafting while ensuring every claim is grounded in actual diff data.
What it does
- Evidence-based drafting converts commit logs, diff hunks, or file lists into structured PR descriptions.
- Risk identification flags migrations, schema changes, and new dependencies found in the code.
- Test plan generation creates actionable verification steps based on the author's work or suggested manual checks.
- Title optimization provides multiple specific title alternatives to replace vague branch names.
- Integrity checking scans for secrets in the diff and ensures no ticket numbers or results are fabricated.
How it works
- Supply the
change_evidencesuch as agit logorgit diff --statalong with an optional intent statement. - The agent scans the files and logic changes to categorize the work by area and impact.
- It generates a structured draft including a summary, why the change matters, and a detailed "What Changed" list.
- Review the generated title alternatives and the pre-open checklist for final manual approval.
Frameworks & tools
This skill is language-agnostic and works with any Git-based workflow, including GitHub, GitLab, and Bitbucket. It interprets standard diff and log outputs.
Why this beats prompting it yourself
Generic prompts often hallucinate ticket numbers or guess the intent behind a change. This skill follows a strict "evidence-first" rule, refusing to invent data and explicitly labeling suggested versus confirmed test plans.
Use cases
- Drafting descriptions for large, complex diffs where manual summarizing is error-prone.
- Standardizing PR quality across a team using a consistent, decision-heavy template.
- Generating clear rollout and risk notes for DevOps and release managers.
Known limitations
Requires raw text input of diffs or commit logs as it cannot directly access private repositories. It will not auto-open or merge PRs.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 8 days ago
- Passed all security checks, Safe to install
Needs access to
Cygnixlabs