StateGraph
StateGraph is a LangGraph class for defining stateful workflows as graphs of nodes, state schemas, and transitions.
Build Python Web Apps with Streamlit
StateGraph is a LangGraph graph-builder API for defining nodes, state, and transitions in a stateful computation graph.
- How it is used
- Builds a stateful, one-node LLM workflow for a chatbot.
- How deep it goes
- Works through state definition, node updates, edges, compilation, and invocation; stops at a single assistant node.
- Examples
- Uses a command-line chatbot graph and a Streamlit single-prompt, single-response chatbot.
- What you can do after
- Build it from scratch: compile and invoke a one-node graph that returns an LLM response.
Read how Build Python Web Apps with Streamlit explains StateGraph →
Knowledge Graphs and LLMs in Action
StateGraph is a LangGraph class for constructing workflows whose nodes operate over shared typed state.
- How it is used
- Constructs a typed, sequential and conditionally routed workflow for a knowledge-graph QA agent.
- How deep it goes
- Defines initialization, node registration, sequential edges, and conditional routing, but does not detail StateGraph internals.
- Examples
- Uses intent detection, schema extraction, text-to-Cypher conversion, query execution, retry routing, and summary generation.
- What you can do after
- Build it from scratch as a named-node pipeline with typed shared state and error-aware conditional execution.
Read how Knowledge Graphs and LLMs in Action explains StateGraph →
Related concepts
- LangGraphused by · LangGraph is a framework for building stateful, graph-structured workflows and applications, including those involving language models and agents.
- agent workflowconfigures · An agent workflow is a structured, multistep process in which one or more AI agents perform or coordinate actions toward an objective, potentially adapting to new information or outcomes.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.