agent-lightning
olmmlo-cmd 제작
Search & Research요약
Microsoft Research's agent training framework.
설명
이 스킬이 하는 일
Agent Lightning is Microsoft Research's framework for training AI agents using Reinforcement Learning, Automatic Prompt Optimization, and Supervised Fine-tuning. It instruments existing agents with minimal code changes via emit helpers or a tracer wrapper. Supports LangChain, AutoGen, CrewAI, and OpenAI Agent SDK out of the box.
Unlike manual prompt engineering or full model retraining, Agent Lightning applies RL and prompt optimization directly to your existing agent code without requiring a rewrite or framework switch.
사용 시점
- Improving a SQL-generation agent's accuracy through RL training
- Tuning system prompts automatically for a customer support agent
- Fine-tuning a coding assistant on preference data
- Running benchmark evaluations on a multi-agent pipeline
- Exporting optimized prompts or model weights for production deployment
예시 워크플로
User asks: train my LangChain agent on SQL generation tasks using reinforcement learning
- 1
Instrument the agent by wrapping it with agentlightning tracer or adding emit_tool_call helpers
- 2
Create a config.yaml specifying agent type as langchain, algorithm as grpo, and eval tasks including sql-generation
- 3
Run
agent-lightning train --config config.yamlto start training - 4
Monitor loss curves and reward metrics via
agent-lightning dashboard --port 8080 - 5
Export the trained prompts or model with
agent-lightning exportfor deployment
Optimized agent with improved SQL generation accuracy, exportable for production use
요구 사항
- OPENAI_API_KEY environment variable
설치
clawhub install olmmlo-cmd/agent-lightningCLI 명령을 실행하거나, 프롬프트를 Claude Code, Cursor, Codex 등 사용하는 코딩 에이전트에 붙여넣으세요.