Use when building search systems that need both semantic similarity and keyword matching - covers combining vector and BM25 search with Reciprocal Rank Fusion, alpha tuning for search weight control, and optimizing retrieval quality
This skill provides complete coverage of Google Gemini embeddings API (gemini-embedding-001) for building RAG systems, semantic search, document clustering, and similarity matching. Use when implementing vector search with Google's embedding models, integrating with Cloudflare Vectorize, or building retrieval-augmented generation systems. Covers SDK usage (@google/genai), fetch-based Workers implementation, batch processing, 8 task types (RETRIEVAL_QUERY, RETRIEVAL_DOCUMENT, SEMANTIC_SIMILARITY, etc.), dimension optimization (128-3072), and cosine similarity calculations. Prevents 8+ embedding-specific errors including dimension mismatches, incorrect task types, rate limiting issues (100 RPM free tier), vector normalization mistakes, text truncation (2,048 token limit), and model version confusion. Includes production-ready RAG patterns with Cloudflare Vectorize integration, chunking strategies, and caching patterns. Token savings: ~60%. Production tested. Keywords: gemini embeddings, gemini-embedding-001, g
Persistent memory compression system for Claude Code enabling context preservation across sessions with automatic observations, semantic search, and privacy controls
Proactive recovery using plan mode and subagents. After 1-2 failed attempts, STOP trying variations. Enter plan mode and launch parallel Explore/Plan agents to find idiomatic solutions instead of spinning wheels.
[1-2 sentence description of what this skill does]. Triggers on [specific phrases/contexts that should activate this skill]. Outputs [what the skill produces].
Build comprehensive randomization lists for creative entropy. Use when you need to create or expand lists of story elements (professions, locations, objects, names, etc.) for use with entropy tools. Leverages research sources like Kiwix/Wikipedia to build lists with good variety and size.
Review and refine MCP skill descriptions to follow the agent skill specification, and record approved overrides in dev-swarm/mcp_descriptions.yaml. Use when the user asks to update one MCP skill description, all skills from a server, or all MCP skills.
Design and orchestrate multi-agent systems. Use when building complex AI systems requiring specialization, parallel processing, or collaborative problem-solving. Covers agent coordination, communication patterns, and task delegation strategies.
Edit images with Gemini using the nanobanana CLI. Use when the user asks to transform or retouch images and wants concrete output files.
Use when wanting to interact with any LLM - Explains available inference endpoints so the agent selects suitable models.
Build an API-oriented agent using AgentRuntime/StreamingRuntime with predictable request/response slices and session support.
Builds LLM applications with LangChain including chains, agents, memory, tools, and RAG pipelines. Use when users request "LangChain setup", "LLM chain", "AI workflow", "conversational AI", or "RAG pipeline".
Discover new Runware AI models from documentation and implement providers. Use when: checking for new Runware models, implementing Runware providers, updating providers.md status, or working with Runware API integrations.
Professional skill creation with research-driven workflow and automated validation. USE WHEN: Creating new skills, validating existing skills, deciding between Skills vs Subagents, migrating documents to skills, or running individual validation tools. PRIMARY TRIGGERS: "create skill" = Full creation (12 steps with research + execution planning) "validate skill" = Validation workflow (steps 3-8) "Skills vs Subagents" = Decision workflow (step 0) "convert doc to skill" = Migration workflow "estimate tokens" = Token optimization "security scan" = Security audit WORKFLOW COMPLIANCE: Structured workflows with validation checkpoints. Research phase (Step 1c-1d) ensures skills based on proven approaches. DIFFERENTIATOR: Research-driven creation. Web search (3-5 queries) before building. Multi-proposal generation. 9 automation scripts. Quality 9.0+/10. REUSED: Anthropic's init_skill.py and package_skill.py (production-tested).
LangChain.js - TypeScript framework for building LLM-powered applications with agents, chains, RAG, tools, memory, and integrations for OpenAI, Anthropic, Google, and hundreds of other providers
Guide for creating effective Claude Code agents. This skill should be used when users want to create a new agent (or update an existing agent) that configures Claude with specialized system prompts, tool restrictions, model selection, and MCP/skill integrations.
Recognize a photo and narrate a kid-friendly explanation using image understanding + TTS.
Create and maintain AgentV YAML evaluation files for testing AI agent performance. Use this skill when creating new eval files, adding eval cases, or configuring custom evaluators (code validators or LLM judges) for agent testing workflows.
