code reviewby rsmdt
Code Quality Review
rsmdt/code-quality-review
Unified code review skill for correctness, design, readability, security, performance, testability, accessibility, and error-handling conventions.
What it changes
Passed 21 of 23 test cases with the skill, and 19 of 23 without it.
Without skill83%
With skill91%
Checks it passed that the model alone missed
- Highlights the SQL injection risk in the query string concatenation.
- Prioritizes the cross-site scripting vulnerability over unused imports.
2 fixed19 passed either way2 still failing0 broke
- Pass-rate lift
- +9 pp
- Tokens
- +38%
- Agent turns
- 0%
- Response time
- +18%
Model gemini-3.6-flashMeasured 8/30/2026v121 of 23 cases passed with skill
See it in action
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptReview the following Node.js Express user search route in `routes/users.js`:
```js
app.get('/users/search', async (req, res) => {
const name = req.query.name;
// Notice inconsistent camelCase naming below
const user_result = await db.query("SELECT * FROM users WHERE username = '" + name + "'");
res.json(user_result.rows);
});
```
Focus on variable naming style and provide code review feedback.
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Highlights the SQL injection risk in the query string concatenation. | Missed | Passed |
|---|
| Provides a parameterized query code fix. | Passed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md59 linessingle file
Persona
Act as a senior reviewer who evaluates code quality holistically and provides prioritized, actionable feedback.
Review Target: $ARGUMENTS
Interface
ReviewFinding {
priority: CRITICAL | HIGH | MEDIUM | LOW
dimension: Correctness | Design | Readability | Security | Performance | Testability | Accessibility | ErrorHandling
title: string
location: string
observation: string
impact: string
suggestion: string
}
State {
target = $ARGUMENTS
findings = []
strengths = []
}
Constraints
Always:
- Prioritize issues that affect correctness, security, and user impact first.
- Include observation, impact, and concrete fix for each finding.
- Verify accessibility and error-handling standards when UI/I/O code is touched.
- Keep feedback constructive and implementation-focused.
Never:
- Focus on stylistic nits over substantive risks.
- Report findings without clear remediation guidance.
- Ignore security/performance/accessibility implications on user-facing paths.
Reference Materials
reference/anti-patterns.md — Common code anti-patterns and remediation strategies
Track record
Benchmark
21 / 23
test cases passed, vs 19 / 23 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026
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
Importedrsmdt / the-startup
plugins/team/skills/quality/code-quality-review/SKILL.md
License MITImported 8/10/2026Version v1