> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omi.me/llms.txt
> Use this file to discover all available pages before exploring further.

# Examples

> Example prompts and integrations with the Omi MCP server

## Example Prompts

Once connected, try these prompts with your AI assistant:

<AccordionGroup>
  <Accordion title="Search and recall" icon="magnifying-glass">
    ```
    "What do you know about my work projects?"
    "Find conversations where I discussed travel plans"
    "Search my memories for anything about fitness goals"
    "What did I talk about last Tuesday?"
    ```
  </Accordion>

  <Accordion title="Memory management" icon="brain">
    ```
    "Remember that I prefer morning meetings before 10am"
    "Update my memory about my favorite restaurant — it's now Sushi Nakazawa"
    "Delete the memory about my old phone number"
    "What are all my memories in the 'work' category?"
    ```
  </Accordion>

  <Accordion title="Conversation analysis" icon="chart-line">
    ```
    "Summarize my conversations from this week"
    "Find the conversation where I brainstormed app ideas"
    "What topics come up most in my recent conversations?"
    "Show me the full transcript of my last meeting"
    ```
  </Accordion>
</AccordionGroup>

***

## Integration Examples

<CardGroup cols={3}>
  <Card title="LangChain" icon="link" href="https://github.com/BasedHardware/omi/tree/main/mcp/examples">
    Build chains with Omi data
  </Card>

  <Card title="OpenAI Agents" icon="robot" href="https://github.com/BasedHardware/omi/tree/main/mcp/examples">
    Create AI agents using Omi
  </Card>

  <Card title="DSPy" icon="code" href="https://github.com/BasedHardware/omi/tree/main/mcp/examples">
    Programmatic LLM pipelines
  </Card>
</CardGroup>

***

## Python SDK Example

This programmatic example uses the manual MCP-key fallback because the adapter receives
static headers. Create an `omi_mcp_...` key as described in the [Setup guide](/doc/developer/mcp/setup)
and load it from a secret store or environment variable in production.

```python theme={null}
import asyncio
import os

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async def main():
    # Current langchain-mcp-adapters API: the client is not a context manager —
    # construct it and await get_tools().
    client = MultiServerMCPClient({
        "omi": {
            "url": "https://api.omi.me/v1/mcp",
            "transport": "streamable_http",
            "headers": {"Authorization": f"Bearer {os.environ['OMI_MCP_API_KEY']}"},
        }
    })
    tools = await client.get_tools()

    agent = create_react_agent(ChatOpenAI(model="gpt-4o-mini"), tools)
    response = await agent.ainvoke({
        "messages": "Search my memories for anything about my morning routine"
    })
    print(response["messages"][-1].content)


if __name__ == "__main__":
    asyncio.run(main())
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.