Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
| 1 | # composio-langgraph |
| 2 | |
| 3 | Adapts Composio tools to LangChain's `StructuredTool` format for use in LangGraph agents and graph workflows, giving them access to 1000+ apps through a single Composio session. |
| 4 | |
| 5 | ## Installation |
| 6 | |
| 7 | ```bash |
| 8 | pip install composio composio-langgraph langgraph langchain langchain-openai |
| 9 | ``` |
| 10 | |
| 11 | Set `COMPOSIO_API_KEY` (get one from [dashboard.composio.dev/settings](https://dashboard.composio.dev/settings)) and `OPENAI_API_KEY` in your environment: |
| 12 | |
| 13 | ```bash |
| 14 | export COMPOSIO_API_KEY=xxxxxxxxx |
| 15 | export OPENAI_API_KEY=xxxxxxxxx |
| 16 | ``` |
| 17 | |
| 18 | ## Quickstart |
| 19 | |
| 20 | Create a session for your user, fetch its tools, and hand them to your agent. The wrapped tools also work anywhere LangGraph accepts LangChain tools, such as a `ToolNode` in a custom graph. |
| 21 | |
| 22 | ```python |
| 23 | from composio import Composio |
| 24 | from composio_langgraph import LanggraphProvider |
| 25 | from langchain.agents import create_agent |
| 26 | from langchain_openai import ChatOpenAI |
| 27 | |
| 28 | composio = Composio(provider=LanggraphProvider()) |
| 29 | llm = ChatOpenAI(model="gpt-5.2") |
| 30 | |
| 31 | # Each session is scoped to one of your users |
| 32 | session = composio.create(user_id="user_123") |
| 33 | tools = session.tools() |
| 34 | |
| 35 | agent = create_agent(tools=tools, model=llm) |
| 36 | result = agent.invoke( |
| 37 | { |
| 38 | "messages": [ |
| 39 | ( |
| 40 | "user", |
| 41 | "Send an email to john@example.com with the subject 'Hello' and body 'Hello from Composio!'", |
| 42 | ) |
| 43 | ] |
| 44 | } |
| 45 | ) |
| 46 | |
| 47 | print(result["messages"][-1].content) |
| 48 | ``` |
| 49 | |
| 50 | ## Error handling |
| 51 | |
| 52 | Each wrapped tool builds its `args_schema` from the Composio tool's input schema. When argument validation fails, the tool does not raise; it returns a structured result: |
| 53 | |
| 54 | ```python |
| 55 | {"successful": False, "error": "<validation message>", "data": None} |
| 56 | ``` |
| 57 | |
| 58 | Check `successful` in tool output instead of wrapping calls in `try`/`except`. |
| 59 | |
| 60 | ## Links |
| 61 | |
| 62 | - [LangChain provider docs](https://docs.composio.dev/docs/providers/langchain) (covers LangGraph) |
| 63 | - [Composio documentation](https://docs.composio.dev) |