Découvrez les skills research llm ai, comparez leurs dépôts et consultez leurs instructions.
Use the ReinforceNow CLI for RLHF training. Use when running rnow commands, initializing projects, submitting training runs, testing rollouts, or downloading models. Triggers on "rnow", "rnow init", "rnow run",...
LLM fine-tuning with LoRA, QLoRA, DPO alignment, and synthetic data generation. Efficient training, preference learning, data creation. Use when customizing models for specific domains.
Local LLM inference with Ollama. Use when setting up local models for development, CI pipelines, or cost reduction. Covers model selection, LangChain integration, and performance tuning.
Testing patterns for LLM-based applications. Use when testing AI/ML integrations, mocking LLM responses, testing async timeouts, or validating structured outputs from LLMs.
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.
Generate training data from teacher models for off-policy distillation. Use when converting prompts to SFT datasets using larger models, generating teacher completions with tools, or preparing data for knowledg...
Write tool functions for ReinforceNow agent training. Use when creating @tool decorated functions, writing tools.py, or sandbox tools. Triggers on "@tool", "tools.py", "tool function", "function calling", "agen...
Write reward functions for ReinforceNow RL training. Use when creating @reward decorated functions, writing rewards.py, using precondition rewards, sandbox rewards, llm_judge, or math-verify. Triggers on "rewar...
Curate and add documents to the golden dataset with multi-agent validation. Use when adding test data, creating golden datasets, saving examples.
Use when designing agent system prompts, optimizing RAG retrieval, or when context is too expensive or slow. Reduces tokens while maintaining quality through strategic positioning and attention-aware design.
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 cho...
Security patterns for LLM integrations including prompt injection defense and hallucination prevention. Use when implementing context separation, validating LLM outputs, or protecting against prompt injection a...
Auto-load relevant memories at session start from both mem0 and graph. Use when you need session context restored or preloaded.
LangGraph Functional API with @entrypoint and @task decorators. Use when building workflows with the modern LangGraph pattern, enabling parallel execution, persistence, and human-in-the-loop.
LLM function calling and tool use patterns. Use when enabling LLMs to call external tools, defining tool schemas, implementing tool execution loops, or getting structured output from LLMs.
LLM streaming response patterns. Use when implementing real-time token streaming, Server-Sent Events for AI responses, or streaming with tool calls.
Agentic workflow patterns for autonomous LLM reasoning. Use when building ReAct agents, implementing reasoning loops, or creating LLMs that plan and execute multi-step tasks.
LangGraph parallel execution patterns. Use when implementing fan-out/fan-in workflows, map-reduce over tasks, or running independent agents concurrently.
Anthropic's Contextual Retrieval technique for improved RAG. Use when chunks lose context during retrieval, implementing hybrid BM25+vector search, or reducing retrieval failures.
Advanced RAG with Self-RAG, Corrective-RAG, and knowledge graphs. Use when building agentic RAG pipelines, adaptive retrieval, or query rewriting.
Comprehensive prompt engineering with Chain-of-Thought, few-shot learning, prompt versioning, and optimization. Use when designing prompts, improving accuracy, managing prompt lifecycle.
Use when conversation context is too long, hitting token limits, or responses are degrading. Compresses history while preserving critical information using anchored summarization and probe-based validation.
Long-term semantic memory across sessions using Mem0. Use when you need to remember, recall, or forget information across sessions, or when referencing what we discussed last time or in a previous session.
LLM observability platform for tracing, evaluation, prompt management, and cost tracking. Use when setting up Langfuse, monitoring LLM costs, tracking token usage, or implementing prompt versioning.
Le classement utilise les étoiles du dépôt GitHub indexé. Plusieurs skills peuvent partager ce nombre : ce n’est ni un nombre d’installations ni une note de qualité.