Claude Forge — a suite of domain-agnostic Claude Code plugins sharing hook infrastructure via symlinks: session continuity, DevOps/AWS/Terraform, AI/LLM patterns, frontend & UI, and engineering & code review. MIT.
Command
Build AI-first applications with RAG pipelines, embeddings, vector databases, agentic workflows, and LLM integration. Use when building an AI feature, integrating an LLM, setting up vector search, or designing agent architectures. Triggers on AI app, LLM integration, RAG, vector database, agentic, prompt engineering…
Command
AI coaching and conversational support — empathetic interactions, safety guardrails, motivational interviewing, SMART goals, and crisis handling for coaching, therapy, education, or customer support. Use when building a coaching or support conversation, adding safety guardrails, or handling sensitive user disclosures.…
Command
Text embeddings for semantic search and similarity. Use when converting text to vectors, choosing embedding models, implementing chunking, or setting up semantic search. Triggers on embeddings, text-embedding, vector, chunking, cosine similarity, semantic search, embedding model, batch embed.
Command
LLM function calling and tool use patterns. Use when enabling LLMs to call external tools, defining tool schemas, building tool execution loops, or returning structured JSON. Triggers on function calling, tool use, tool schema, structured output, Pydantic validation, parallel tool calls, bind tools.
Command
Quality criteria, workflows, and multi-agent analysis for curating high-quality golden dataset entries. Use when curating golden dataset entries, defining quality criteria, or reviewing candidate examples. Triggers on golden dataset curation, quality criteria, dataset entry review, example selection, curate dataset.
Command
Backup, restore, and validation of golden datasets for AI/ML systems, protecting test data against catastrophic loss. Use when backing up a golden dataset, restoring from backup, or guarding against data loss. Triggers on golden dataset backup, restore dataset, dataset management, test data integrity, data loss…
Command
Validation rules, schema checks, duplicate detection, and coverage analysis for golden dataset integrity. Use when validating a golden dataset, checking schema conformance, finding duplicate entries, or measuring coverage. Triggers on golden dataset validation, schema check, duplicate detection, coverage analysis…
Command
LLM observability with self-hosted Langfuse — tracing, evaluation, prompt versioning, cost tracking, and LLM-as-judge scoring. Use when adding LLM tracing, monitoring token cost, versioning prompts, or building evaluation dashboards. Triggers on Langfuse, LLM observability, tracing, prompt versioning, cost tracking…
Command
LangGraph checkpointing and persistence. Use when implementing fault-tolerant workflows, resuming interrupted executions, or debugging with state history. Triggers on LangGraph checkpoint, persistence, fault tolerance, resume workflow, state history, checkpointer, thread state.
Command
LangGraph human-in-the-loop patterns. Use when implementing approval workflows, manual review gates, user feedback integration, or interactive agent steps. Triggers on human-in-loop, approval workflow, review gate, interrupt, user feedback, manual approval, interactive agent.
Command
LangGraph parallel execution patterns. Use when implementing fan-out/fan-in workflows, map-reduce over tasks, or running independent agents concurrently. Triggers on LangGraph parallel, fan-out, fan-in, map-reduce, concurrent agents, parallel branches, Send API.
Command
LangGraph conditional routing patterns. Use when implementing dynamic routing based on state, creating branching workflows, or building conditional retry loops. Triggers on LangGraph routing, conditional edge, dynamic routing, branching workflow, retry loop, state-based routing.
Command
LangGraph state management patterns. Use when designing workflow state schemas, choosing TypedDict vs Pydantic, or accumulating state with reducers. Triggers on LangGraph state, state schema, TypedDict, Pydantic state, reducer, Annotated, accumulating state, addmessages.
Command
LangGraph supervisor-worker pattern. Use when building a coordinator agent that routes to specialized workers, or implementing round-robin or priority delegation. Triggers on LangGraph supervisor, worker pattern, coordinator agent, agent delegation, round-robin, task routing, orchestrator.
Command
LLM output evaluation and quality assessment. Use when implementing LLM-as-judge patterns, adding quality gates for AI outputs, or building automated evaluation pipelines. Triggers on LLM evaluation, LLM-as-judge, quality gate, output scoring, hallucination detection, eval pipeline, multi-dimension scoring.
Command
LLM streaming response patterns. Use when implementing real-time token streaming, Server-Sent Events for AI responses, or streaming with tool calls. Triggers on LLM streaming, token streaming, SSE, streaming response, stream tool calls, real-time LLM, incremental output.
Command
Testing patterns for LLM-based applications. Use when testing AI/ML integrations, mocking LLM responses, handling async timeouts, or validating structured output. Triggers on LLM testing, mock LLM, VCR, test AI, async timeout, structured output validation, deterministic LLM test.
Command
Multi-agent fan-out/fan-in orchestration and synthesis. Use when coordinating multiple specialized agents, delegating via a supervisor, resolving conflicting agent outputs, or synthesizing parallel research. Triggers on multi-agent, orchestration, fan-out, fan-in, supervisor delegation, agent coordination, synthesis…
Command
Google NotebookLM programmatic access — create notebooks, manage sources (URLs, PDFs, YouTube, Drive), generate audio/video/slides/quizzes, and chat with documents. Use when building a research notebook, generating a podcast from docs, or querying a source collection. Triggers on notebooklm, podcast from docs…
Command
Local LLM inference with Ollama. Use when setting up local models for development, running models in CI pipelines, or reducing inference cost. Triggers on Ollama, local LLM, local inference, offline model, self-hosted model, LangChain Ollama, model quantization.
Command
Production hybrid search with PGVector and BM25 using Reciprocal Rank Fusion, metadata filtering, and performance tuning for semantic retrieval. Use when building hybrid semantic and keyword search, tuning PGVector performance, or filtering by metadata. Triggers on pgvector, hybrid search, BM25, reciprocal rank…
Command
Provider-native prompt caching for Claude and OpenAI. Use when optimizing LLM costs with cache breakpoints, caching system prompts, or reducing token cost for repeated prefixes. Triggers on prompt caching, cache breakpoint, cachecontrol, token cost, cached prefix, system prompt caching.
Command
RAG pipeline patterns for grounded LLM responses. Use when building a Q&A system, adding citations, implementing a knowledge base, or preventing hallucinations. Triggers on RAG, retrieval augmented, knowledge base, Q&A pipeline, citations, hybrid search, context retrieval, hallucination prevention.
Command
Redis semantic caching for LLM applications. Use when caching LLM responses by vector similarity, cutting cost on repeated queries, or building a multi-level cache. Triggers on semantic cache, Redis cache, vector similarity cache, LLM cache, response caching, cache warming, TTL cache.