LangChain and LangGraph - Failproof AI

Three steps to your first LangChain or LangGraph trace. For streaming, human-in-the-loop, span naming, and options, see the full guide.

Install

pip install 'failproofai-sdk[langgraph]'

For LangChain without LangGraph, use failproofai-sdk[langchain].

Instrument

import failproofai_sdk

failproofai_sdk.configure(environment="production")
failproofai_sdk.instrument()

with failproofai_sdk.session():
    graph.invoke({"messages": [HumanMessage("...")]})

No call site changes. instrument() registers a tracer that LangChain injects into every callback manager it builds, so graphs, tools, and models are captured — including ones inside libraries you did not write.

Check it arrived

Run it once, then open Observe → Sessions and select your environment. Your run appears as a reconstructed trace. If nothing arrives, confirm the machine is connected with failproofai config --status.

Full LangChain guide

What gets recorded, streaming, token counts, span naming, session control, options, and human in the loop.