Claude Skills

Biome 2.0+ linting and formatting for fast, unified code quality. Includes type inference, ESLint migration, CI integration, and 421 lint rules. Use when migrating from ESLint/Prettier or setting up new projects.

Anthropic's Contextual Retrieval technique for improved RAG. Use when chunks lose context during retrieval, implementing hybrid BM25+vector search, or reducing retrieval failures.

Text embeddings for semantic search and similarity. Use when converting text to vectors, choosing embedding models, implementing chunking strategies, or building document similarity features.

HyDE (Hypothetical Document Embeddings) for improved semantic retrieval. Use when queries don't match document vocabulary, retrieval quality is poor, or implementing advanced RAG patterns.

CLIP, SigLIP 2, Voyage multimodal-3 patterns for image+text retrieval, cross-modal search, and multimodal document chunking. Use when building RAG with images, implementing visual search, or hybrid retrieval.

Production hybrid search combining PGVector HNSW with BM25 using Reciprocal Rank Fusion. Use when implementing hybrid search, semantic + keyword retrieval, vector search optimization, metadata filtering, or choosing between HNSW and IVFFlat indexes.

Query decomposition for multi-concept retrieval. Use when handling complex queries spanning multiple topics, implementing multi-hop retrieval, or improving coverage for compound questions.

Retrieval-Augmented Generation patterns for grounded LLM responses. Use when building RAG pipelines, constructing context from retrieved documents, adding citations, or implementing hybrid search.

Use when backing up, restoring, or validating golden datasets. Prevents data loss and ensures test data integrity for AI/ML evaluation systems.

Use when validating golden dataset quality. Runs schema checks, duplicate detection, and coverage analysis to ensure dataset integrity for AI evaluation.

LLM output evaluation and quality assessment. Use when implementing LLM-as-judge patterns, quality gates for AI outputs, or automated evaluation pipelines.

Orchestrate single user-invocable skill across 3 parallel scenarios with synchronized state and progressive difficulty. Use for demos, testing, and progressive validation workflows.

Graph-first memory orchestration - knowledge graph (PRIMARY, always available) with optional mem0 cloud enhancement for semantic search. Use when designing memory orchestration or combining graph and mem0.

Message queue patterns with RabbitMQ, Redis Streams, and Kafka. Use when implementing async communication, pub/sub systems, event-driven microservices, or reliable message delivery.

End-to-end testing with Playwright 1.57+. Use when testing critical user journeys, browser automation, cross-browser testing, AI-assisted test generation, or validating complete application flows.

[BUILD] Full-power feature implementation with parallel subagents. Use when implementing, building, or creating features.

Fix GitHub issue with parallel analysis and implementation. Use when fixing issues, resolving bugs, closing GitHub issues.

Enforces FastAPI Clean Architecture with blocking validation. Use when implementing router-service-repository patterns, enforcing layer separation, or validating dependency injection in backend code.

Advanced Celery patterns including canvas workflows, priority queues, rate limiting, multi-queue routing, and production monitoring. Use when implementing complex task orchestration, task prioritization, or enterprise-grade background processing.