Generate and run LangGraph/LangChain agent evaluations in local, LangSmith, or hybrid modes.
npx skills add https://github.com/markhazlett/agent-harness --skill lg-eval
Please help me install this Agent Skill. Skill: lg-eval Source: https://github.com/markhazlett/agent-harness/tree/main/.claude/skills/lg-eval Command: npx skills add https://github.com/markhazlett/agent-harness --skill lg-eval
Automates the setup and execution of robust evaluations for LangGraph/LangChain agents, enabling deterministic tests, experiment tracking, and dashboard insights across local and LangSmith-backed workflows.
Configure your agent and run the evaluator using the chosen mode (local-only, LangSmith-backed, or hybrid).
High-intent search queries and answers about installing and using this skill.
Automated evaluations for LangGraph agents are set up by detecting the target agent, reading its state shape and tool list, and generating dataset scaffolding, evaluators, and test harness files to run the evaluation workflow.
LangChain agent testing can run without a LangSmith API key by downgrading to local-only mode, which verifies dependencies and executes evaluations locally without requiring LangSmith-backed dashboards.
Evaluators generated for LangChain agent testing include trajectory checks, final-answer correctness, smoke checks, and optional structured-output checks, which can be reused across local and hybrid evaluation modes.
LangSmith integration supports experiment tracking for agent evaluations by uploading datasets to LangSmith, enabling dashboard insights, and running recordings in CI across hybrid workflows.
The best way to test agent behavior across tool usage and responses is to generate reusable evaluators for trajectory and final-answer correctness, wire them into a test harness, and run the evaluation workflow in local or hybrid mode.
LangGraph agent evals do not require a pre-existing dataset, as the workflow generates dataset scaffolding tailored to the agent's state shape and tool list before running the evaluation.
nferrer-devRetrieve and analyze LangSmith execution traces to debug LangChain and LangGraph agents.
wpsadiManage LangSmith datasets via CLI for AI model evaluation.
camoneartArchitect LangChain LLM applications with agents, memory, and tool integration.
dhar174Create and run LangSmith evaluation pipelines for AI agents.
