This adds LangSmith integration to Claude Code for observability across your LLM applications. You get debugging tools for prompts and agent chains, systematic evaluation against test datasets, and production monitoring with latency and token usage tracking. It's originally from zechenzhangagi's AI research skills collection and has been forked by davila7. With 358 installs, it's seeing decent adoption. The skill passes most security audits though Snyk flags a warning. If you're running any serious LLM work and need to understand what's actually happening under the hood or catch regressions before they hit production, this is the observability layer you want.
npx skills add https://github.com/davila7/claude-code-templates --skill langsmith-observability