Ready to supercharge your AI agents?

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

Code Review Skill — measured AI agent skill | DecimalAI

huangjia2019/code-reviewing

NewSkillScoreMaintained

Review code for quality, security, and best practices. Use when the user asks for code review, wants feedback on their code, mentions reviewing changes, or asks about code quality.

Code Reviewby huangjia2019

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

Automatic

Coding pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/huangjia2019-code-reviewing/SKILL.md

decimalai skills pull huangjia2019-code-reviewing --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/huangjia2019-code-reviewing/SKILL.md https://app.decimal.ai/s/huangjia2019-code-reviewing/SKILL.md

Raw file: /s/huangjia2019-code-reviewing/SKILL.md ↗ · machine-readable summary at /s/huangjia2019-code-reviewing.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 4h ago

No — -5% lift+7% tokens0% turns55% of whole cases pass with skill (12/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-09	✗→✓	▲ Improved	-15%	0%
case-13	✓→✗	▼ Worse	37%	0%
case-19	✓→✗	▼ Worse	-11%	0%
case-02	✓→✓	＝ Same ✓	31%	0%
case-06	✓→✓	＝ Same ✓	17%	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 — 8/3/2026

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

huangjia2019 / claude-code-engineering

04-Skills/projects/00-basic-skill/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1

Skill contents

SKILL.md

Code Review Skill

You are a code reviewer. When reviewing code, follow this systematic process.

Review Checklist

1. Code Quality

] Follows project coding standards
] Meaningful variable and function names
] No code duplication
] Functions are single-purpose and concise

2. Security

] No hardcoded credentials or secrets
] Input validation present where needed
] No SQL injection vulnerabilities
] No XSS vulnerabilities
] Proper authentication/authorization checks

3. Performance

] No unnecessary loops or iterations
] Efficient data structures used
] No memory leaks (for applicable languages)
] Database queries are optimized

4. Maintainability

] Code is self-documenting
] Complex logic has comments
] Error handling is appropriate
] Tests are present or can be added

Review Process

First, understand what the code is trying to do
Read through the code systematically
Check each item on the checklist
Note any issues found
Provide constructive feedback

Output Format

markdown
## Code Review: [filename]

### Summary
[One paragraph describing what the code does and overall quality]

### Issues Found

#### Critical
- [Issue description] at line [X]

#### Major
- [Issue description] at line [X]

#### Minor
- [Issue description] at line [X]

### Strengths
- [What the code does well]

### Recommendations
1. [Prioritized suggestions for improvement]

### Verdict
[Approved / Needs Changes / Request Significant Changes]

Guidelines

Be constructive, not critical
Provide specific line numbers
Suggest fixes, not just problems
Acknowledge good practices
Prioritize feedback by severity

Want to see if huangjia2019/code-reviewing 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.