עבריתLog in

Integrations

Connect OneLi to agent frameworks and automation tools.

Choosing a connection

  • MCP with an API key: most frameworks connect to the OneLi MCP server with Authorization: Bearer oli_… and get every tool, including events, consultations and instructions.
  • Your own REST tool: when you want full control over the fields, or no MCP. See the call lifecycle and the generic recipe.
The examples were checked against each package’s documentation on 22 October 2026, at the versions listed on each page. Model ids in the examples are placeholders: use the model you work with.

Agent frameworks

MastraMastra has a built-in MCP client. Add the OneLi server with your API key in a header and the agent gets every OneLi tool.OpenAI Agents SDKIn the OpenAI Agents SDK, connect OneLi as a Streamable HTTP MCP server with your API key in a header. There is also a hosted variant where OpenAI calls the server, and a plain function tool over the REST API.Vercel AI SDKIn AI SDK v7, create an MCP client with createMCPClient from @ai-sdk/mcp and pass its tools to generateText or an agent.LangChain / LangGraphSince langchain 1.4 the docs present the built-in MCPAdapter (beta) as the replacement for MultiServerMCPClient. Agents made with create_agent run on LangGraph.Claude Agent SDK / Anthropic APIIn the Claude Agent SDK, configure OneLi as an http MCP server with an Authorization header and allow its tools. With the Anthropic Messages API you can use the MCP connector, which calls the server itself.Google ADKIn Google ADK, add an McpToolset with StreamableHTTPConnectionParams and an Authorization header to the agent. A plain Python function also works as a tool.CrewAIIn CrewAI, pass an MCPServerHTTP with an Authorization header to the agent through mcps.LlamaIndexIn LlamaIndex, create a BasicMCPClient with an Authorization header, turn it into a tool list with McpToolSpec and pass it to a FunctionAgent.Pydantic AIIn Pydantic AI v2, use MCPToolset, which infers the transport, and pass it to the agent’s toolsets.

Automation tools

n8nn8n offers two paths: an MCP Client Tool node attached to an AI Agent node, which gets every OneLi tool, or an HTTP Request node that creates a call task directly.Make / ZapierIn Make and Zapier, the tested path is an HTTP request to the REST API with the key in a header. Both also have an MCP Client app in beta, which we have not tested yet.