AgentSkillsCN

Agent 9 skill(CodeRabbitReviewer)

代理 9 技能（CodeRabbitReviewer）

SKILL.md

Skill: CodeRabbit Reviewer

Name

CodeRabbit Reviewer - Automated Code Review Request & Integration

Description

This skill automates the triggering and management of CodeRabbit AI code reviews using the configuration from /home/fitna/homelab/.coderabbit.yaml. It initiates focused reviews on pull requests, applies homelab-specific review rules (3-system deployment, agent prompts, infrastructure configs), validates deployment scripts for security, and tracks review findings for Sprint 26 tasks.

When to Use This Skill

Trigger Conditions

Use this skill when the user requests ANY of the following:

  • •"Run CodeRabbit review on [PR/file]"
  • •"Request code review"
  • •"Check this code with CodeRabbit"
  • •"Validate [code/config] for deployment"
  • •"Review agent prompt for Sprint 26"
  • •"Run security check on [file]"

Context Indicators

  • •User mentions pull requests or code review
  • •User wants automated security/quality checks
  • •User discusses CodeRabbit or AI code review
  • •User needs validation before merging PR
  • •User references .coderabbit.yaml configuration

Process Steps

Phase 1: Review Configuration Verification (2 minutes)

  1. •

    Check CodeRabbit Configuration Reference: /home/fitna/homelab/.coderabbit.yaml

    Key Settings:

    yaml
    reviews:
  auto_review:
    enabled: true
    drafts: false  # Don't review draft PRs

  focus_areas:
    - architecture
    - security
    - performance
    - best_practices
    - maintainability

  path_instructions:
    - path: "infrastructure/**"
      instructions: |
        Review for:
        - Multi-system deployment compatibility (VPS, ThinkPad, RTX1080)
        - Docker Compose syntax and best practices
        - Secret management (no hardcoded credentials)
        - Health check implementations
        - Rollback strategy documentation

    - path: "agents/**"
      instructions: |
        Review for:
        - Agent prompt structure (80-120 lines)
        - Shared context references
        - Measurable success criteria
        - Output path specifications

    - path: "sprints/sprint-26/**"
      instructions: |
        Review for Sprint 26 specific requirements:
        - Task allocation across 6 agents
        - Deliverable file paths in prompts
        - Brand guidelines compliance
  2. •

    Verify GitHub CLI Authentication

    bash
    gh auth status

# If not authenticated
gh auth login
  3. •

    Check Repository Connection

    bash
    gh repo view fitna/homelab --web

Phase 2: Trigger Review (3-5 minutes)

  1. •

    Review Pull Request

    Automatic Review (on PR creation):

    • •CodeRabbit automatically reviews when PR created (if not draft)
    • •Check PR for CodeRabbit comment

    Manual Review Trigger:

    bash
    # Using coderabbit CLI
coderabbit review --pr PR_NUMBER

# Or via gh CLI (trigger re-review)
gh pr comment PR_NUMBER --body "@coderabbitai review"
  2. •

    Review Specific Files

    bash
    # Review single file
coderabbit review --file path/to/file.py

# Review directory
coderabbit review --path infrastructure/docker/stacks/

# Review with focus area
coderabbit review --pr PR_NUMBER --focus security
  3. •

    Apply Homelab-Specific Checklist

    Infrastructure Files:

    • • No secrets in Docker Compose files
    • • All services have health checks
    • • Environment variables use .env references
    • • Traefik labels configured correctly
    • • Volume mounts use named volumes (not bind mounts)
    • • Network mode appropriate for service
    • • Restart policy set (unless-stopped)

    Deployment Scripts:

    • • SSH commands use key authentication (no passwords)
    • • Error handling implemented (set -e)
    • • Idempotent (safe to run multiple times)
    • • Logging to deployment log directory
    • • Health checks before proceeding
    • • Rollback strategy documented
    • • No hardcoded IP addresses (use variables)

    Agent Prompts (Sprint 26):

    • • Metadata section complete (agent, model, priority, deadline, output_path)
    • • Shared context references present (@project-context.md, etc.)
    • • Objective is 1-2 sentences and measurable
    • • 5-10 requirements listed
    • • Exact deliverable file paths specified
    • • 3-5 success criteria (checkboxes)
    • • Copy-pasteable execution block included
    • • Prompt length 80-120 lines

    Python Code (AI Agents Platform):

    • • Type hints on all functions
    • • FastAPI endpoint decorators correct
    • • Pydantic models for validation
    • • Async/await used correctly
    • • Error handling with try/except
    • • Logging statements for debugging
    • • No SQL injection vulnerabilities
    • • Authentication/authorization checks

Phase 3: Review Analysis (5 minutes)

  1. •

    Read CodeRabbit Findings

    bash
    # View PR comments
gh pr view PR_NUMBER --comments

# Or via web
gh pr view PR_NUMBER --web
  2. •

    Categorize Findings

    Critical (Must Fix):

    • •Security vulnerabilities (SQL injection, XSS, hardcoded secrets)
    • •Breaking changes without migration path
    • •Missing error handling for critical operations
    • •Deployment scripts without rollback

    High (Should Fix):

    • •Performance issues (N+1 queries, missing indices)
    • •Missing health checks
    • •Incorrect Docker Compose syntax
    • •Missing environment variable validation

    Medium (Consider Fixing):

    • •Code duplication (DRY violations)
    • •Missing type hints
    • •Suboptimal algorithms
    • •Missing logging statements

    Low (Nice to Have):

    • •Minor style issues
    • •Documentation improvements
    • •Refactoring opportunities
  3. •

    Generate Summary Report

    markdown
    ## CodeRabbit Review Summary - PR #XX

**Files Reviewed:** X
**Findings:** X critical, X high, X medium, X low

### Critical Issues
- [File:Line] Security: Hardcoded API key in config
- [File:Line] Security: Missing input validation (SQL injection risk)

### High Priority
- [File:Line] Missing health check in docker-compose.yml
- [File:Line] No error handling in deployment script

### Recommendations
1. Fix critical security issues before merging
2. Add health checks to all services
3. Implement error handling in scripts
4. Consider refactoring duplicated code in [files]

**Merge Recommendation:** ❌ Do not merge until critical issues resolved

Phase 4: Address Findings (Variable)

  1. •

    Fix Critical Issues

    • •Address all critical and high findings
    • •Update code based on recommendations
    • •Commit fixes to PR branch
  2. •

    Request Re-Review

    bash
    # After fixes, request re-review
gh pr comment PR_NUMBER --body "@coderabbitai review"

# Or push commits (auto-triggers review)
git push origin feature-branch
  3. •

    Track Review Resolution

    • • All critical issues resolved
    • • High priority issues addressed or documented as tech debt
    • • CodeRabbit approves changes
    • • Tests passing (if applicable)

Phase 5: Sprint 26 Integration (Continuous)

  1. •

    Apply Sprint 26 Validation Rules Reference: .coderabbit.yaml Sprint 26 section

    Agent Prompt Validation:

    • •Verify 80-120 line structure
    • •Check shared context references
    • •Validate deliverable paths exist
    • •Confirm success criteria are measurable

    Infrastructure Config Validation:

    • •Verify 3-system compatibility (VPS, ThinkPad, RTX1080)
    • •Check Docker Compose version compatibility
    • •Validate secret references (no hardcoded values)
    • •Confirm health checks implemented
  2. •

    Update Sprint Board with Review Status Update /home/fitna/homelab/sprints/sprint-26/SPRINT_BOARD.md:

    markdown
    - [x] [Agent] Task description (X SP) | Priority: Y | Owner: Z | Review: ✅ Approved

Rules and Constraints

Hard Rules (Must Follow)

  1. •NEVER merge PR with critical security findings
  2. •ALWAYS review infrastructure changes (deployment risk)
  3. •Agent prompts MUST pass 80-120 line validation
  4. •Deployment scripts MUST have rollback strategy
  5. •Secrets MUST NEVER be hardcoded in any file
  6. •ALL critical and high findings MUST be addressed before merge

Soft Rules (Best Practices)

  • •Request review before marking PR ready
  • •Address medium priority findings if time permits
  • •Document technical debt for low priority issues
  • •Use CodeRabbit suggestions as learning opportunities

Quality Gates

Before approving PR for merge:

  • • CodeRabbit review completed
  • • 0 critical findings
  • • 0 high priority findings (or documented exceptions)
  • • Tests passing (if applicable)
  • • Documentation updated
  • • Sprint 26 rules validated (if applicable)

Expected Outputs

Deliverables

  1. •Review Summary Report - Categorized findings
  2. •Issue Tracking - Critical/high issues logged
  3. •Merge Recommendation - Approve or block
  4. •Sprint Board Update - Review status on tasks

Success Metrics

  • •✅ CodeRabbit review triggered successfully
  • •✅ All findings categorized by severity
  • •✅ Critical issues resolved before merge
  • •✅ Review turnaround time <24 hours

Error Handling

Common Issues & Solutions

Issue: CodeRabbit not reviewing PR

  • •Solution: Check .coderabbit.yaml for drafts: false setting
  • •Action: Mark PR as ready for review (not draft)

Issue: GitHub CLI not authenticated

  • •Solution: Run gh auth login and follow prompts
  • •Check: gh auth status

Issue: Review findings unclear

  • •Solution: Ask CodeRabbit for clarification
  • •Command: gh pr comment PR_NUMBER --body "@coderabbitai explain [finding]"

Issue: False positive findings

  • •Solution: Add suppression comment or update .coderabbit.yaml
  • •Action: Document reason for suppression

Issue: Review takes too long

  • •Solution: Review specific files/paths instead of full PR
  • •Command: coderabbit review --path specific/path/

Example Usage

User Request

"Run CodeRabbit review on PR #42 for the new monitoring stack deployment."

Agent Response

markdown
I'll trigger CodeRabbit review for PR #42 (monitoring stack deployment).

**Triggering Review...**
```bash
gh pr comment 42 --body "@coderabbitai review"

Review Configuration Applied:

  • •Focus areas: architecture, security, performance, best_practices
  • •Path-specific rules: infrastructure/** (multi-system deployment validation)

Review Completed! Summary:

Files Reviewed: 3

  • •infrastructure/docker/stacks/monitoring.yml
  • •infrastructure/docker/prometheus/prometheus.yml
  • •infrastructure/docker/grafana/provisioning/datasources/prometheus.yml

Findings:

  • •0 critical ✅
  • •1 high ⚠️
  • •2 medium ℹ️
  • •3 low 💡

High Priority Issue: ⚠️ monitoring.yml:45 - Grafana admin password hardcoded Recommendation: Move to .env file and reference via ${GRAFANA_ADMIN_PASSWORD}

Medium Priority: ℹ️ prometheus.yml:12 - Scrape interval 5s may cause high load Recommendation: Increase to 15s for production

ℹ️ monitoring.yml:78 - Missing health check for Loki service Recommendation: Add healthcheck section

Merge Recommendation: ⚠️ Fix high priority issue (hardcoded password) before merging

Next Steps:

  1. •Move Grafana password to .env file
  2. •Consider increasing scrape interval to 15s
  3. •Add health check for Loki
  4. •Request re-review after fixes
code

## Integration Points

### Related Skills
- **Agent 2 - DeploymentOrchestrator**: Validate deployment scripts before execution
- **Agent 3 - AgentTaskGenerator**: Validate generated agent prompts
- **Agent 8 - SOPExecutor**: Review SOP documentation for accuracy

### External Tools
- GitHub CLI (`gh`)
- CodeRabbit CLI (`coderabbit`)
- Configuration: `/home/fitna/homelab/.coderabbit.yaml`

### Data Sources
- CodeRabbit config: `/home/fitna/homelab/.coderabbit.yaml`
- Deployment prompt: `/home/fitna/homelab/CODERABBIT_DEPLOYMENT_PROMPT.md`
- Sprint board: `/home/fitna/homelab/sprints/sprint-26/SPRINT_BOARD.md`

## Version
v1.0 - Initial skill definition based on CodeRabbit integration analysis