{"aiPlatform":"claude-code@2025.06","category":"development","commandName":"/ai-review","content":"---\nname: AI/ML Code Review\ndescription: Specialized code review tool for AI/ML projects, focusing on model quality, data handling, LLM implementation, and production readiness. Analyzes reproducibility, prompt injection prevention, privacy compliance, and performance optimization.\nallowed_tools:\n  - filesystem      # Code file analysis\n  - memory          # Track review patterns\n  - sqlite          # Store review metrics\ntags:\n  - ai-ml\n  - code-review\n  - model-validation\n  - data-science\n  - llm-optimization\ncategory: development\nversion: 1.0.0\nauthor: AI Commands Team\n---\n\n# AI/ML Code Review\n\nPerform a specialized AI/ML code review for: $ARGUMENTS\n\nConduct comprehensive review focusing on:\n\n1. **Model Code Quality**:\n   - Reproducibility checks\n   - Random seed management\n   - Data leakage detection\n   - Train/test split validation\n   - Feature engineering clarity\n\n2. **AI Best Practices**:\n   - Prompt injection prevention\n   - Token limit handling\n   - Cost optimization\n   - Fallback strategies\n   - Timeout management\n\n3. **Data Handling**:\n   - Privacy compliance (PII handling)\n   - Data versioning\n   - Preprocessing consistency\n   - Batch processing efficiency\n   - Memory optimization\n\n4. **Model Management**:\n   - Version control for models\n   - A/B testing setup\n   - Rollback capabilities\n   - Performance benchmarks\n   - Drift detection\n\n5. **LLM-Specific Checks**:\n   - Context window management\n   - Prompt template security\n   - Response validation\n   - Streaming implementation\n   - Rate limit handling\n\n6. **Vector Database Review**:\n   - Embedding consistency\n   - Index optimization\n   - Query performance\n   - Metadata management\n   - Backup strategies\n\n7. **Production Readiness**:\n   - GPU/CPU optimization\n   - Batching strategies\n   - Caching implementation\n   - Monitoring hooks\n   - Error recovery\n\n8. **Testing Coverage**:\n   - Unit tests for preprocessing\n   - Integration tests for pipelines\n   - Model performance tests\n   - Edge case handling\n   - Mocked LLM responses\n\nProvide specific recommendations with severity levels (Critical/High/Medium/Low). Include code examples for improvements and links to relevant best practices.","contentHash":"efab38f191345b94b0e834924e8112a2b2e8569149df300e054a496d6fab3f4b","copies":0,"createdAt":"2025-08-12T16:09:35.416Z","description":"Specialized review for AI/ML codebases","github":{"repoUrl":"https://github.com/Commands-com/commands","lastSyncDirection":"from-github","metadata":{"importedFrom":"github_repository","repoPrivate":false,"repoDefaultBranch":"main","connectedAt":"2025-08-12T16:09:35.416Z"},"importedAt":"2025-08-12T16:09:35.416Z","lastSyncAt":"2025-08-17T17:57:45.376Z","fileMapping":{"license":null,"readme":null,"assets":[],"mainFile":"tools/ai-review.md"},"selectedCommand":"ai-review","fileShas":{"mainFile":"90333e95d31faaa2c0caf553ab5caab3841e708e","yamlPath":"313647b1fb381389da33b7913e95baf617c4b392"},"branch":"main","connectionType":"commands_yaml","connected":true,"lastSyncCommit":"01591bc061d236bde47bf23b0f47e8afcf1a5144","importSource":"repository_import","installationId":"69232615","syncStatus":"synced"},"githubRepoUrl":"https://github.com/Commands-com/commands","id":"f0dfc3f9-efe7-4b43-bc69-53b648d09c1d","instructions":"Specialized review for AI/ML codebases","likes":0,"mcp_search_content":"","organizationUsername":"commands-com","price":"free","search_content":"ai review specialized review for ai/ml codebases /ai-review development claude-code@2025.06","title":"AI Review","type":"command","updatedAt":"2025-08-17T17:57:45.376Z","userId":"W0V8NAw5AhWRwcuwSoFLOi1Yem83","visibility":"public","name":"ai-review","userInteraction":{"userHasStarred":false}}