Ready to supercharge your AI agents?

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

PR Review — measured AI agent skill | DecimalAI

choutos/pr-review

SkillScore95Maintained

Review pull requests for code quality, correctness, and maintainability. Use when asked to review a PR, check a diff, or provide code feedback. Triggers on: review, PR, pull request, code review, diff, merge request.

by choutos

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

decimalai skills pull choutos-pr-review

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/choutos-pr-review/SKILL.md

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

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 3d ago

Yes — +18% lift-8% tokens0% turns95% of whole cases pass with skill (21/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-03	✗→✓	▲ Improved	-24%	0%
case-04	✗→✓	▲ Improved	-18%	0%
case-10	✗→✓	▲ Improved	70%	0%
case-13	✗→✓	▲ Improved	-49%	0%
case-02	✓→✓	＝ Same ✓	-22%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/2/2026

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

choutos / agent-skills-spec

examples/pr-review/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1

Skill contents

SKILL.md

PR Review

Process

Read the PR description and linked issue (if any)
Review the diff file by file
Provide feedback privately first (never post to GitHub/GitLab without explicit approval)

What to Check

Correctness

Does the code do what the PR claims?
Are edge cases handled?
Could this break existing functionality?

Quality

Clear naming and structure?
Appropriate error handling?
Tests added or updated?
No hardcoded values that should be config?

Security

No secrets in code?
Input validation present?
SQL injection / XSS risks?

Operations

Will this affect deployment?
Database migrations needed?
Config changes required?

Feedback Format

For each issue found:

**[severity]** file:line — description

Suggestion: ...

Severity levels: critical (blocks merge), important (should fix), nit (optional improvement).

Rules

Never post reviews publicly without explicit approval
Acknowledge what's done well, not just problems
Be specific: "line 42 has X" beats "the code has issues"
Suggest fixes, don't just point out problems

Want to see if choutos/pr-review would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.