wernerhe/super-model/super-code-review
- community
Checking that the SKILL.md file is still in the repository.
Run a comprehensive code review at a milestone (typically before merging a feature branch). Writes an HMAC-signed cache marker for super-prepare-branch to consume; if nothing changed since the last review, super-prepare-branch can skip re-running.
Install
npx vskill@latest install wernerhe/super-model/super-code-reviewSKILL.md
super-code-review
The comprehensive code review skill. Catches the issues that fast feedback loops miss - dead code, namespace drift, dependency rot, coverage gaps - by sweeping the diff systematically across 9 dimensions.
When to use
- A feature branch is "done" and the user is about to push / open PR / merge.
- A risky refactor needs a once-over before going downstream.
- A
super-prepare-branchinvocation finds no cache marker and chain-fires this skill.
The 9-module sweep
file-references- dead imports, orphan files, broken doc links.namespace-consistency- new identifiers fit existing naming conventions.dead-code- functions / variables / imports defined but never used in the diff.test-coverage- changed lines covered by tests; new branches have tests.security-scan- secrets in plaintext, command injection, dependency vulns.dependency-health- new deps vetted, version pins reasonable, no abandoned packages.doc-drift- public API changes reflected in docs / READMEs / docstrings.performance-hotspots- O(n^2) inside loops, accidental N+1, repeated regex compiles.complexity- functions over a threshold (cyclomatic / cognitive), single-responsibility violations.
Each module produces zero or more findings tagged with severity (critical / important / minor) and an optional fix proposal.
Cache marker output
On successful completion, writes:
via super_lib.cache.write_marker. The payload includes:
verdict: READY-TO-MERGE | NEEDS-FIXES.findings: list of findings with severity, file, line range, description, suggested fix.commit_sha: HEAD SHA at review time.timestamp: UTC ISO-8601.
HMAC-signed so a downstream super-prepare-branch can trust the verdict; tampered markers are treated as missing and trigger a re-review.
Hard rules
- NEVER claim "READY-TO-MERGE" with unresolved CRITICAL findings.
- ALWAYS write the cache marker, even on NEEDS-FIXES verdict - super-prepare-branch needs the record.
- NEVER auto-apply fixes from modules marked
fix_policy: never. Always present and ask.
Verify after running
<project>/.super/cache/code-review-<sha>.jsonexists with a valid HMAC.- The verdict matches the user's assessment (no over-stating "ready").
- Each finding has a clear file:line and a description the user can act on.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/super-code-review/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/wernerhe/super-model/super-code-review)