# Process Code Review Findings

Code review documents can leave teams with a long, hard-to-track fix list. This skill turns those findings into an ordered workflow for fixes, skips, checks, and status updates.

## Install

```bash
npx skillstore add diversioteam/process-code-review
```

## Metadata

- Status: approved
- Slug: diversioteam-process-code-review
- Skillstore revision: r1
- Version status: missing
- Tree hash: cf88980091bed43be9828ddb974f83bb126ddc74d37d73ec683e38c9d7fd6374
- Author: DiversioTeam
- GitHub username: DiversioTeam
- License: MIT
- Repository: https://github.com/DiversioTeam/agent-skills-marketplace/tree/main/plugins/process-code-review/skills/process-code-review
- Ref: c579efd02184f1128d6741d16e6c8d951d7d4208
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 70
- Public page: https://skillstore.pages.dev/skills/diversioteam-process-code-review
- Manifest: https://skillstore.pages.dev/api/skills/diversioteam-process-code-review/manifest

## Capabilities

- Parses review documents with BLOCKING, SHOULD\_FIX, and NIT severity tags.
- Creates a TodoWrite list that tracks each issue through completion.
- Presents findings in priority order for fix or skip decisions.
- Applies accepted fixes and updates review documents with status markers.
- Runs documented checks such as ruff, ty, and Django checks when applicable.

## Use Cases

- Resolve Generated Review Findings: Work through a generated review document, fix accepted issues, and keep the review file updated.
- Triage Team Review Debt: Review a teammate review file, skip unsuitable fixes with reasons, and preserve an audit trail.
- Verify Blocking Issues: Focus on high-priority findings first and confirm that security or correctness problems are addressed.

## Prompt Templates

### Process a Review File

```
Use process-code-review on docs/code_reviews/service_review.md and walk me through each finding interactively.
```

### Preview Before Editing

```
Run process-code-review in dry-run mode on api_review.md. Summarize findings by severity without changing files.
```

### Focus on Blocking Issues

```
Process only BLOCKING findings in payments_review.md. Ask before each fix and update the review document status.
```

### Automate Trusted Cleanup

```
For a trusted generated review file, process NIT findings automatically, run available checks, and summarize modified files.
```

## Limitations

- Designed for review documents that follow the expected Markdown structure.
- Requires project tools such as ruff or ty before related checks can run.
- Auto mode can change many files and should be limited to trusted review documents.
- Does not commit changes; a separate pre-commit workflow is expected.

## Best Practices

- Use interactive mode for unfamiliar or externally supplied review documents.
- Read source context before accepting proposed fixes from the review document.
- Run quality gates after each meaningful change and record skipped reasons.

## Anti Patterns

- Running auto mode on untrusted review documents.
- Treating proposed fixes as authoritative without checking current source code.
- Marking issues fixed before tests or relevant checks are complete.

## Security Audit

- Audited at: 2026-07-05T13:33:45.269\+00:00
- Summary: All 42 static findings are false positives caused by Markdown backticks, code fences, slash-command examples, and documented quality-check commands. No prompt injection, secret access, network exfiltration, or hidden executable code was found in SKILL.md. The main residual risk is workflow-level: auto mode can apply changes from untrusted review documents.

## Stats

- Views: 176
- Downloads: 10
- Favorites: 0
- Popularity score: 0