Ready to supercharge your AI agents?

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

Review a pull request — measured AI agent skill | DecimalAI

superman2003/review-pull-request

SkillScore95Maintained

Use when asked to review a pull request, diff, or branch changes. Provides a methodical review checklist covering correctness, security, and maintainability.

by superman2003

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

Automatic

Coding pack

decimalai skills pull superman2003-review-pull-request

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/superman2003-review-pull-request/SKILL.md

Raw file: /s/superman2003-review-pull-request/SKILL.md ↗ · machine-readable summary at /s/superman2003-review-pull-request.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 4d ago

Yes — +14% lift-28% 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-05	✗→✓	▲ Improved	-35%	0%
case-06	✗→✓	▲ Improved	-11%	0%
case-16	✗→✓	▲ Improved	-45%	0%
case-01	✓→✗	▼ Worse	-66%	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

superman2003 / agentpack

examples/code-review-toolkit/skills/review-pull-request/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1Bundled 1 file

Skill contents

references/

SKILL.md

Review a pull request

Review changes in this order, and report findings grouped by severity.

Instructions

Read the full diff before commenting on anything.
Check correctness first: logic errors, off-by-one, unhandled error paths, race conditions.
Check security: injection, secrets in code, unsafe deserialization, missing input validation.
Check tests: does new behavior have coverage? Do the tests assert outcomes, not implementation?
Check maintainability last: naming, duplication, dead code.
See references/checklist.md for the full item-by-item checklist.

Reporting

Group findings as: blocking, should-fix, nit.
Quote the exact file and line for every finding.
If the change looks good, say so plainly; do not invent nits.

Want to see if superman2003/review-pull-request 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.