⚠️ — This project is in early development. APIs and skill content may change.
Agent skills for observing and evaluating LLM applications with LangSmith. Query traces, build evaluation datasets, and create custom evaluators — all from your coding agent.
These skills can be installed for any agent supported by skills.sh, including Claude Code, Deep Agents CLI, Cursor, Windsurf, Goose, and many more.
Using npx skills:
Local (current project):
npx skills add langchain-ai/langsmith-skills --skill '*' --yesGlobal (all projects):
npx skills add langchain-ai/langsmith-skills --skill '*' --yes --globalTo link skills to a specific agent (e.g. Claude Code):
npx skills add langchain-ai/langsmith-skills --agent claude-code --skill '*' --yes --globalInstall directly as a Claude Code plugin:
/plugin marketplace add langchain-ai/langsmith-skills
/plugin install langsmith-skills@langsmith-skillsAlternatively, clone the repo and use the install script:
# Install for Claude Code in current directory (default)
./install.sh
# Install for Claude Code globally
./install.sh --global
# Install for DeepAgents CLI in current directory
./install.sh --deepagents
# Install for DeepAgents CLI globally (includes agent persona)
./install.sh --deepagents --global| Flag | Description |
|---|---|
--claude |
Install for Claude Code (default) |
--deepagents |
Install for DeepAgents CLI |
--global, -g |
Install globally instead of current directory |
--force, -f |
Overwrite skills with same names as this package |
--yes, -y |
Skip confirmation prompts |
After installation, set your API keys:
export LANGSMITH_API_KEY=<your-key>
export OPENAI_API_KEY=<your-key> # For OpenAI models
export ANTHROPIC_API_KEY=<your-key> # For Anthropic modelsThen run your coding agent from the directory where you installed (for local installs) or from anywhere (for global installs).
- langsmith-trace - Query and export traces (includes helper scripts)
- langsmith-dataset - Generate evaluation datasets from traces (includes helper scripts)
- langsmith-evaluator - Create custom evaluators (includes helper scripts)
Note: All skills include Python and TypeScript helper scripts for common operations.
Agent configuration lives in config/. To update an existing installation:
./install.sh --force