AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
예측 불가능한 AI 에이전트를 시뮬레이션·평가·관측으로 신뢰성 있는 운영 시스템으로 전환
AI agent testing and evaluation that turns unpredictable agents into reliable production systems, with simulations, evals, observability, and governance.
개발 도구TypeScript · GherkinGitHub @langwatch온라인운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
The platform for LLM evaluations and AI agent testing
저장소 README에서 가져옴
README 전체를 불러오는 중입니다. 아래는 저장된 발췌입니다.
LangWatch is the open-source platform for AI in production: trace, test, route and govern every LLM call in your company, from the agents you build to the coding assistants your engineers use. LangWatch is Apache 2.0. View docs.
Only Node.js required. For production, see Self-hosting LangWatch.README 전문 보기 ↗LangWatch is the open-source platform for AI in production: trace, test, route and govern every LLM call in your company, from the agents you build to the coding assistants your engineers use. LangWatch is Apache 2.0. View docs.
npx langwatch claude # or codex, copilot, opencode, ...
npx @langwatch/server
Only Node.js 24 or newer required (macOS or glibc Linux). For production, see Self-hosting LangWatch.
Coding assistants: Claude Code · Codex · GitHub Copilot · opencode · Cursor · MCP server · more
Frameworks: LangChain · LangGraph · Vercel AI SDK · Mastra · CrewAI · Google ADK · DSPy · OpenTelemetry · more
Model providers: OpenAI · Anthropic · Azure OpenAI · Vertex AI · Bedrock · more
Governance sources: Copilot Studio · Claude Cowork · Anthropic and OpenAI compliance APIs · Workato · Databricks Genie · OpenTelemetry and S3 audit feeds · more
No-code platforms: LangFlow · Flowise · n8n · more
Read CONTRIBUTING.md, then make quickstart. Bugs and feature requests go in GitHub issues, questions in Discord.
Apache 2.0. The enterprise modules under enterprise/ need a commercial license in production, and the SDKs are MIT. See Editions and licensing and LICENSE.md.
GDPR compliant and ISO 27001 certified. Report vulnerabilities to [email protected].