Stats
Actions
Tags
From cadre-devkit-claude
Reviews unstaged and staged git changes for code quality, security, style, error handling, best practices, and test coverage. Outputs structured assessment: APPROVE/REQUEST CHANGES, positives, required fixes with file:line refs, suggestions, security concerns.
How this command is triggered — by the user, by Claude, or both
Slash command
/cadre-devkit-claude:reviewThis command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# Review Command **Purpose:** Qualitative code review - checks code quality, style, security, and best practices. **Distinct from /validate:** This command does human-like code review. Use `/validate` for automated checks (types, lint, tests, build). **Workflow:** `/review` → `/validate` → `/ship` ## Process ### 1. Gather Changes Get the current diff to review: !`git diff --name-only` !`git diff --staged --name-only` If there are changes, get the full diff: !`git diff` !`git diff --staged` If no changes found, inform the user and stop. ### 2. Perform Review Use the code-reviewer...
Purpose: Qualitative code review - checks code quality, style, security, and best practices.
Distinct from /validate: This command does human-like code review. Use /validate for automated checks (types, lint, tests, build).
Workflow: /review → /validate → /ship
Get the current diff to review:
!git diff --name-only
!git diff --staged --name-only
If there are changes, get the full diff:
!git diff
!git diff --staged
If no changes found, inform the user and stop.
Use the code-reviewer agent for comprehensive review. The agent has skills: code-formatter, error-handler which auto-load.
Task(
subagent_type="code-reviewer",
prompt="Review the following code changes for quality, security, and best practices.
Files changed:
[list files from step 1]
Review focus:
- Code quality and maintainability (DRY, KISS, YAGNI)
- Security vulnerabilities (comprehensive OWASP checklist)
- Style consistency (via code-formatter skill)
- Error handling patterns (via error-handler skill)
- Test coverage for new/changed code
Provide a structured review with:
- Overall assessment (APPROVE / REQUEST CHANGES)
- Positive findings
- Required changes (with file:line references)
- Suggestions for improvement
- Security concerns (if any)"
)
Present the agent's review to the user with clear next steps.
After review is complete:
/validate to verify automated checks, then /ship to commit/reviewUser: /review
Claude:
1. Gathers git diff (3 files changed)
2. Spawns code-reviewer agent with diff context
3. Agent reviews using its skills and security checklist
4. Presents structured review:
## Code Review Summary
**Overall Assessment**: APPROVE
### Positive Findings
- Clean separation of concerns in new utility functions
- Good error handling with descriptive messages
### Suggestions
- `src/utils/parser.ts:45` - Consider extracting the regex to a named constant
### Security Concerns
- None identified
---
Ready for `/validate` → `/ship`
npx claudepluginhub benshapyro/cadre-devkit-claude/reviewAnalyzes local Git diffs and runs a thorough code review, producing a structured report with critical, warning, and suggestion categories.
/code-reviewReviews uncommitted code changes via git diff for security vulnerabilities like hardcoded secrets and SQL injection, plus quality issues, generating a severity-based report with line numbers, descriptions, and fixes. Blocks commits on critical/high issues.
/code-reviewReviews uncommitted changes via git diff for security vulnerabilities, code quality issues, and best practices; generates severity-based report with fixes and blocks commits on critical/high issues.
/reviewReviews staged changes or recent commits across five axes—correctness, readability, architecture, security, performance—producing categorized findings with file:line references and fixes.
/reviewRuns Codex code review on local git state (working tree or vs base branch). Supports --wait/--background, --base <ref>, --scope auto|working-tree|branch.
/reviewInvokes multiple external AI CLIs (Gemini, Claude, Codex, etc.) to independently review a phase plan and produces a structured REVIEWS.md with per-reviewer feedback.