Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

pr-description · jnmetacode | DecimalAI

code reviewby jnmetacode

pr-description

jnmetacode/pr-description

Source jnmetacode/skillet ↗License MIT

Write a reviewer-first pull-request description from the actual branch diff.

SkillScore77Maintained

Automatic

Coding pack

Last verified 8/3/2026

What it changes

Passed 17 of 22 test cases with the skill, and 13 of 22 without it.

Without skill59%

With skill77%

Checks it passed that the model alone missed

The response includes a dedicated risk section.
The response explicitly categorizes changed items by intent type.
Formatting changes are explicitly categorized as mechanical churn.

and 4 more in the full report

5 fixed12 passed either way4 still failing1 broke

Pass-rate lift: +18 pp
Tokens: -5%
Agent turns: 0%
Response time: -33%

Model gemini-3.6-flashMeasured 8/3/2026v117 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/jnmetacode-pr-description/SKILL.md https://app.decimal.ai/s/jnmetacode-pr-description/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add jnmetacode-pr-description, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/jnmetacode-pr-description/ and pins it in .decimal/skills.lock

decimalai skills add jnmetacode-pr-description

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/jnmetacode-pr-description/SKILL.md https://app.decimal.ai/s/jnmetacode-pr-description/SKILL.md

Raw file: /s/jnmetacode-pr-description/SKILL.md ↗ · machine-readable summary at /s/jnmetacode-pr-description.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 5

Prompt

I spent 3 days discussing redis caching strategies in Slack before making a simple diff in `cache.go` setting TTL to 300 seconds for repository `acme/cache-service`. Write the opening sentence for the pull request.

What was checked, without and with the skill
Checked	Without skill	With skill
The opening states the concrete problem without summarizing past Slack conversation history.	Missed	Passed

case-19 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md43 linessingle file

pr-description

The reader is a reviewer deciding how to review, not a historian. Write from the diff, not from memory of the conversation.

Procedure

Establish the real change set: git log main..HEAD --oneline and git diff main...HEAD --stat. Read the diff of every file a reviewer will question.
Open with one sentence of why — the problem or goal, not the activity ("Checkout fails for SSO users" beats "Update auth flow").
What changed, grouped by intent rather than by file: behavior changes first, then refactors, then mechanical churn (renames, formatting) called out explicitly so reviewers can skim past it.
How to verify: the exact commands/steps you ran and their result. If CI covers it, say which jobs; if manual, give the click-path.
Risk and rollout: what could break, feature flags, migrations, revert plan. One line each. Omit the section only if genuinely none.
Call out anything you want focused review on ("the locking in store.js is the risky part") — reviewers allocate attention where the author points.

Anti-patterns

Restating the commit list — the PR already shows it.
"See ticket" as the why — summarize the ticket in one line, then link it.
Screenshots without a caption saying what to look at.
Describing the diff file-by-file — group by intent.

Skeleton

Fixes checkout failing for SSO users (#412).

**What**
- Treat IdP-initiated sessions as authenticated in the cart guard (behavior)
- Extract session checks into `session.ts` (refactor, no behavior change)

**Verify**: `npm test` (all green) + manual SSO login → add to cart → checkout.

Track record

Benchmark

17 / 22

test cases passed, vs 13 / 22 without

Adoption

Users: 2
Times used: 50

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

jnmetacode / skillet

skills/pr-description/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

pr-description

Procedure

Anti-patterns

Skeleton

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills