> For the complete documentation index, see [llms.txt](https://ottodata.gitbook.io/ottodata-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ottodata.gitbook.io/ottodata-docs/mcp-for-ai-agents/overview.md).

# Overview

## MCP for AI agents

Otto ships as a **hosted MCP server** — the first narrative stream on Robinhood Chain that AI agents can plug into directly.

### Endpoint

```
https://api.ottodata.app/mcp
```

* Transport: **streamable HTTP** (stateless)
* Authentication: **none** — free during beta
* Tools: **17**, all read-only (every tool carries `readOnlyHint` annotations)

### What agents can do with it

Once connected, an agent can answer questions like:

* *"Which Hood token has the biggest overnight premium right now?"* → `hood_gaps_live`
* *"Is any pool price diverging from its oracle?"* → `oracle_deviation`
* *"Were there large NVDA mints in the last 48 hours?"* → `token_flows`
* *"What's Reddit watching that also shows an on-chain dislocation?"* → `hood_attention`
* *"Any depeg risk on USDG?"* → `stable_risk`

The agent chains tools itself — basis, pools, flows, social, news and deployments are all one call away.

### Design notes

* Tools are thin wrappers over the [REST API](https://ottodata.gitbook.io/ottodata-docs/api-reference/overview); when hosted, calls run in-process (no extra network hop).
* Everything is a query — no tool writes, trades, signs or spends anything. Safe to enable without confirmation prompts.
* The server is also published as a manifest (`server.json`) for the official MCP Registry.

→ [Install in your client](https://ottodata.gitbook.io/ottodata-docs/mcp-for-ai-agents/install-in-your-client) · [Tool reference](https://ottodata.gitbook.io/ottodata-docs/mcp-for-ai-agents/tool-reference)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://ottodata.gitbook.io/ottodata-docs/mcp-for-ai-agents/overview.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.