Track LLM calls, token usage, and costs in LangChain/LangGraph applications.
npx skills add https://github.com/eco2-team/backend --skill langsmith-telemetry
Please help me install this Agent Skill. Skill: langsmith-telemetry Source: https://github.com/eco2-team/backend/tree/main/.claude/skills/langsmith-telemetry Command: npx skills add https://github.com/eco2-team/backend --skill langsmith-telemetry
LangSmith telemetry provides end-to-end observability for LangChain/LangGraph apps by tracking LLM calls, token usage, and costs, enabling teams to monitor performance and spending.
Enable LangSmith tracing by setting LANGCHAIN_TRACING_V2=true and LANGCHAIN_API_KEY, and instrument your workflow with traceable decorators and a LangSmith run config.
High-intent search queries and answers about installing and using this skill.
Enable LangSmith tracing by setting LANGCHAIN_TRACING_V2=true and configuring your LANGCHAIN_API_KEY, then instrument workflows with traceable decorators to capture LLM calls and metadata.
Track token usage and costs in multi-agent pipelines by attaching RunTree metadata to LLM calls, allowing you to monitor spending and performance across tool calls and agent steps.
LangSmith supports optional OpenTelemetry integration, allowing you to unify observability dashboards across development and production environments for comprehensive LLM workflow monitoring.
Tracing tool calls and LLM responses in LangGraph workflows is supported end-to-end, capturing full execution paths across multi-agent pipelines for detailed performance analysis.
A LangSmith API key is required to enable observability, as it authenticates your application to send tracing data, token usage metrics, and cost estimates to the LangSmith platform.
RunTree metadata captures token usage data across LLM calls and tool executions, enabling accurate cost estimates and detailed spending analysis for multi-agent and pipeline workflows.
Harmeet10000Instrument and query LangSmith traces in Python and TypeScript projects.
ContextJet-aiAdd OpenTelemetry tracing to LLM applications capturing prompts, tool calls, token usage, latency, and cost.
hoodiniBuild end-to-end LLM applications with LangChain and LangGraph.
sammcjGuide AWS Strands Agents SDK and AgentCore deployments across Lambda, ECS/Fargate, and AgentCore Runtime.
