Skip to main content

​
Overview

The Omi MCP server enables AI assistants like Claude to interact with your Omi data using natural language. Query your memories, manage conversations, and automate workflows through AI-powered tools.

Read Memories

Retrieve and search your memories

Manage Data

Create, edit, and delete memories

Access Conversations

Browse full conversation transcripts

​
Hosted MCP Server

The easiest way to specific Omi as an MCP server is to use our hosted endpoint with SSE (Server-Sent Events).

Generate an API Key

Open the Omi app and navigate to Settings → Developer → MCP to generate your API key.

Configure your Client

Use the following connection details in your MCP client (like Poke):
  • Server URL: https://api.omi.me/v1/mcp/sse (or shown custom URL)
  • API Key: omi_mcp_... (your generated key)

​
Example: Poke

Poke MCP Setup

​
Using Docker MCP Server

If you prefer to run the MCP server locally:

Generate an API Key

Open the Omi app and navigate to Settings → Developer → MCP to generate your API key.

Install Docker

Install Docker to run the MCP server. We recommend OrbStack for macOS.

Configure Claude Desktop

Add the following to your claude_desktop_config.json:
"mcpServers": {
  "omi": {
    "command": "docker",
    "args": ["run", "--rm", "-i", "-e", "OMI_API_KEY=your_api_key_here", "omiai/mcp-server"]
  }
}
Replace your_api_key_here with the key you generated in Step 1.
The API key can also be provided with each tool call. If not provided, the server uses the OMI_API_KEY environment variable as a fallback.

​
Available Tools

Retrieve a list of user memories.Parameters:
NameTypeRequiredDescription
limitnumberNoMaximum number of memories to retrieve (default: 100)
categoriesarrayNoCategories to filter by (default: [])
Returns: JSON object containing list of memories
Create a new memory.Parameters:
NameTypeRequiredDescription
contentstringYesContent of the memory
categoryMemoryFilterOptionsYesCategory of the memory
Returns: Created memory object
Edit an existing memory’s content.Parameters:
NameTypeRequiredDescription
memory_idstringYesID of the memory to edit
contentstringYesNew content for the memory
Returns: Status of the operation
Delete a memory by ID.Parameters:
NameTypeRequiredDescription
memory_idstringYesID of the memory to delete
Returns: Status of the operation
Retrieve a list of user conversations.Parameters:
NameTypeRequiredDescription
include_discardedbooleanNoInclude discarded conversations (default: false)
limitnumberNoMaximum number of conversations (default: 25)
Returns: List of conversation objects containing transcripts, timestamps, geolocation, and structured summaries

​
Example Integrations

LangChain

Build chains with Omi data

OpenAI Agents

Create AI agents using Omi

DSPy

Programmatic LLM pipelines

​
Debugging

Use the MCP inspector to debug the server:
npx @modelcontextprotocol/inspector uvx mcp-server-omi
For local development:
cd path/to/servers/src/omi
npx @modelcontextprotocol/inspector uv run mcp-server-omi

​
Advanced Configuration

​
Custom Backend URL

If you are self-hosting the Omi backend, specify the API endpoint by setting the OMI_API_BASE_URL environment variable: 
export OMI_API_BASE_URL="https://your-backend-url.com"
This is only needed for self-hosted Omi instances. The default URL points to the official Omi API.

​
Comparison with Developer API

FeatureMCPDeveloper API
PurposeAI assistant integrationDirect HTTP API access
AccessRead/write with AI contextRead & write user data
Use CaseClaude Desktop, AI agentsCustom apps, dashboards
Best ForAI-powered workflowsBatch operations, integrations
For programmatic access without AI assistants, use the Developer API instead.

Developer API

Direct HTTP API for programmatic access

GitHub Examples

LangChain, OpenAI Agents, and DSPy examples

Discord Community

Get help from the community

Contribution Guide

Contribute to the MCP server