Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →langchain-ai/writing-pr-body-title-description
Use when writing or editing a pull request title or body.
Passed 22 of 23 test cases with the skill, and 10 of 23 without it.
and 12 more in the full report
13 fixed9 passed either way0 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
A newer attempt on gemini-3.8-flash on 10/21/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
I fixed the chart legend alignment issue in `components/AnalyticsChart.tsx` where long labels overlapped chart lines. Uploaded before and after image links to the issue ticket. Draft the PR title and description.
| Checked | Without skill | With skill |
|---|---|---|
| The response uses a before-and-after table to show visual changes. | Missed | Passed |
| The response excludes testing checklist or validation sections. | Missed | Passed |
The instructions your model reads when the skill loads.
Keep PR titles and bodies concise; do not write essays. Focus on bullet points, Mermaid diagrams, and useful code samples or snippets. Do not include test-running or validation sections. Do not narrate intermediate PR details; describe the final squash-merge outcome. For visual changes, include a before/after table with uploaded images or videos. In GitHub PR descriptions, always embed screenshots using explicit HTML <img> tags with an explicit width (for example, width="640"), including inside before/after table cells; never use Markdown image syntax. Preserve exact signed image URLs and follow the shared screenshot guidance for published-body, rendered-image, and anonymous GET verification. For benchmarks, always include a before/after table. For truly impressive changes, use a technical-blog style while remaining focused.
test cases passed, vs 10 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.