Skip to main content
Agent framework integrations hook into framework callbacks, processors, or middleware to trace agent lifecycles and model operations: LLM calls, tool invocations, sub-agent handoffs, guardrails, and related steps. Use the table below to choose a supported SDK, then open the integration guide for framework-specific setup. Shared options are session_id, user_id, tags, and metadata in Python, and sessionId, userId, tags, and metadata in TypeScript. Vercel middleware currently accepts an optional span name; use the TypeScript session() and user() scopes for attribution.
All integrations automatically produce properly nested span hierarchies. You do not need to create traces or spans manually.

LangGraph

Callback handler for LangGraph graphs and ReAct agents

LangChain

Callback handler for create_agent and LCEL pipelines

DeepAgents

Callback handler for deep agents with sub-agent dispatch

Google ADK

Monkey-patch instrumentation for ADK runners and tools

Claude Agent SDK

Python adapter and TypeScript query wrapper

CrewAI

Crew kickoff, agents, LLM, and tool spans

OpenAI Agents SDK

First-class TracingProcessor integration

Vercel AI SDK

TypeScript middleware for generateText and streamText