Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Standalone code review methodology for structured, severity-classified code assessment
.claude/skills/josstei-code-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 47% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 27% | 0% |
Activate this skill when performing standalone code reviews via the runtime-specific Maestro review entrypoint or during orchestration quality gates (post-phase checks and final completion gate). This skill provides the methodology for scoping, executing, and reporting code reviews.
Determine review scope using the following priority order:
glob tool to resolve them to concrete file paths before delegating to the code-reviewer agentgit diff --staged produces output, review staged changesgit diff HEAD~1Always confirm the detected scope with the user before proceeding.
If scope is provided as file paths and a git diff is empty for some paths (for example, new unstaged files), include those files' current contents directly in review context so they are still reviewed.
code-reviewer agent with:When delegating to the code-reviewer agent, include:
Issues that could cause security vulnerabilities, data loss, or system crashes:
Issues that cause bugs, design flaws, or significant maintainability problems:
Issues related to style, naming, or minor convention violations:
Optional improvements that enhance readability or maintainability:
Present findings in a structured table followed by a summary:
## Code Review Results
**Scope**: [description of what was reviewed]
**Files Reviewed**: [count]
**Total Findings**: [count by severity]
### Findings
| # | Severity | File | Line | Description | Suggested Fix |
|---|----------|------|------|-------------|---------------|
| 1 | Critical | path/to/file.ts | 42 | [description] | [fix] |
| 2 | Major | path/to/file.ts | 87 | [description] | [fix] |
### Summary
[1-2 paragraph summary of overall code quality, patterns observed, and priority actions]Every finding must:
Do NOT report:
Calibrate the depth and focus of review based on the type of change being reviewed:
When reviewing a diff that contains multiple change types (new files + modifications + deletions), apply the appropriate calibration to each file independently. Do not apply "new file" depth to a file that only had a minor modification.
When reviewing multiple files, identify and consolidate findings that share the same root cause.
### Systemic Finding: [Pattern Violation Name]
- **Severity**: [Critical | Major | Minor | Suggestion]
- **Description**: [What the pattern violation is and why it matters]
- **Affected Locations**:
- `path/to/file1.ext:line` — [brief context]
- `path/to/file2.ext:line` — [brief context]
- `path/to/file3.ext:line` — [brief context]
- **Remediation**: [Single recommendation that addresses all instances]This produces cleaner, more actionable review output by surfacing systemic issues as patterns rather than repeating the same finding across multiple files.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 2,769 | 8,973 | +224% | 1 | 1 | 0% | 391 | 1,668 | +327% | 0 | 0 | — |
case-02 | fail→fail | 14,513 | 16,913 | +17% | 1 | 1 | 0% | 1,537 | 3,063 | +99% | 0 | 0 | — |
case-03 | fail→fail | 13,996 | 3,846 | -73% | 1 | 1 | 0% | 768 | 1,645 | +114% | 0 | 0 | — |
case-04 | pass→pass | 8,112 | 14,977 | +85% | 1 | 1 | 0% | 1,463 | 3,263 | +123% | 0 | 0 | — |
case-05 | pass→fail | 13,011 | 5,388 | -59% | 1 | 1 | 0% | 2,439 | 1,597 | -35% | 0 | 0 | — |
case-06 | pass→pass | 10,443 | 11,015 | +5% | 1 | 1 | 0% | 1,591 | 2,979 | +87% | 0 | 0 | — |
case-07 | fail→pass | 6,725 | 2,363 | -65% | 1 | 1 | 0% | 1,212 | 1,759 | +45% | 0 | 0 | — |
case-08 | fail→fail | 10,428 | 5,803 | -44% | 1 | 1 | 0% | 1,601 | 2,280 | +42% | 0 | 0 | — |
case-22 | pass→pass | 10,260 | 2,658 | -74% | 1 | 1 | 0% | 1,507 | 1,764 | +17% | 0 | 0 | — |
case-09 | pass→pass | 9,891 | 1,618 | -84% | 1 | 1 | 0% | 1,385 | 1,612 | +16% | 0 | 0 | — |
case-10 | pass→pass | 14,679 | 7,837 | -47% | 1 | 1 | 0% | 2,518 | 2,768 | +10% | 0 | 0 | — |
case-11 | pass→pass | 7,873 | 3,633 | -54% | 1 | 1 | 0% | 1,269 | 1,955 | +54% | 0 | 0 | — |
case-12 | pass→pass | 10,095 | 4,565 | -55% | 1 | 1 | 0% | 1,820 | 2,073 | +14% | 0 | 0 | — |
case-13 | fail→pass | 8,817 | 3,603 | -59% | 1 | 1 | 0% | 1,295 | 1,958 | +51% | 0 | 0 | — |
case-14 | pass→pass | 6,785 | 2,560 | -62% | 1 | 1 | 0% | 1,019 | 1,793 | +76% | 0 | 0 | — |
case-15 | pass→pass | 5,758 | 5,003 | -13% | 1 | 1 | 0% | 837 | 2,168 | +159% | 0 | 0 | — |
case-16 | fail→pass | 10,505 | 5,157 | -51% | 1 | 1 | 0% | 1,470 | 2,161 | +47% | 0 | 0 | — |
case-17 | pass→pass | 7,463 | 1,957 | -74% | 1 | 1 | 0% | 1,160 | 1,624 | +40% | 0 | 0 | — |
case-18 | fail→pass | 13,517 | 4,435 | -67% | 1 | 1 | 0% | 2,128 | 1,997 | -6% | 0 | 0 | — |
case-19 | pass→pass | 8,745 | 1,984 | -77% | 1 | 1 | 0% | 1,248 | 1,641 | +31% | 0 | 0 | — |
case-20 | fail→pass | 14,681 | 8,823 | -40% | 1 | 1 | 0% | 2,126 | 2,691 | +27% | 0 | 0 | — |
case-21 | pass→pass | 11,889 | 5,081 | -57% | 1 | 1 | 0% | 1,834 | 2,273 | +24% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 19 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 19 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.