Show HN: Evals Skills

Category: ai-ml

Tags: llm, observability, evals, agent-testing, ai-gateway

Score: 7.5/10 (Innovation: 6, Technical: 7, Documentation: 9, Utility: 8)

LangWatch is an open-source observability and evaluation platform for LLM-powered applications and agents. It provides tracing, testing, routing, and governance across the entire AI stack, from custom agents to coding assistants, with a particular focus on evals and simulation testing. The Skills Directory extends this by offering reusable, CLI-driven agent skills that help developers instrument their code and manage platform resources programmatically.

Target audience: AI engineers, ML platform teams, backend developers building LLM agents

Repository: https://langwatch.ai/docs/skills/directory · TypeScript · Apache-2.0 · 4986 stars

View on Hacker News