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

# Overview

Give AI assistants and agents access to Nansen onchain data. Use a connector, or connect to Nansen MCP manually with an API key.

Nansen gives AI assistants read access to its onchain data: labeled wallets, Smart Money activity, token flows, wallet PnL, Hyperliquid perps, and prediction markets. Everything is served over the [Model Context Protocol](https://modelcontextprotocol.io) (MCP), so any MCP-compatible client can use it.

There are two ways to connect. Most people should start with a connector.

|            | [Connectors](/mcp/connectors.md) (recommended)                         | [Manually connecting to Nansen MCP](/mcp/connecting.md)                     |
| ---------- | ---------------------------------------------------------------------- | --------------------------------------------------------------------------- |
| Best for   | Chatting with Nansen data in ChatGPT, Claude, Grok, or Muse            | Coding agents, IDEs, automation, and custom agents                          |
| Clients    | ChatGPT, Claude (web, desktop, mobile, Claude Code, Slack), Grok, Muse | Claude Desktop, Claude Code, Cursor, VS Code, n8n, and any other MCP client |
| Sign-in    | Your Nansen account, through OAuth                                     | A Nansen API key in a request header                                        |
| Tools      | [12 curated tools](/mcp/connectors.md#curated-tools)                   | [Full tool catalog](/mcp/connecting/tools.md) (50 tools)                    |
| Server URL | `https://mcp.nansen.ai/ra/connector-curated/mcp`                       | `https://mcp.nansen.ai/ra/mcp`                                              |
| Setup      | Install from a directory or follow the custom connector guide          | Paste the server URL and your API key into the client's configuration       |

Both paths are read-only: neither can trade, sign transactions, or move assets. Both bill tool calls to the [Nansen API credits](/getting-started/credits.md) of the account you connect, not to your AI subscription. One prompt can trigger several tool calls, and each tool has its own credit cost. See [credits per call](/mcp/connecting/tools.md#quick-reference).

### Connectors

Install Nansen from your AI client's directory or follow its custom connector guide. You sign in with your Nansen account, approve read-only access, and start asking questions. There is no API key to create, paste, or rotate.

* [ChatGPT](/mcp/connectors/chatgpt.md): install from the ChatGPT plugin directory
* [Claude](/mcp/connectors/claude.md): install from the Claude connector directory. The same sign-in works in [Claude Code](/mcp/connectors/claude.md#claude-code) and in Slack through [Claude Tag](/mcp/connectors/claude-tag.md)
* [Grok](/mcp/connectors/grok.md): add Nansen as a custom connector
* [Muse](/mcp/connectors/muse.md): paste a setup prompt and let Muse configure a custom connector

Every connector exposes the same [12 curated tools](/mcp/connectors.md#curated-tools). They cover the most common research tasks: search, token info and prices, quant scores, token screening, token flows, buyers and sellers, Smart Money holdings, wallet portfolios, counterparties, related wallets, and wallet PnL.

### Manually connecting to Nansen MCP

Connect directly to the Nansen MCP server with a Nansen API key when you need more than the curated set, or when your client has no Nansen connector. This gives you the [full tool catalog](/mcp/connecting/tools.md): everything in the curated set, plus Smart Money DEX trades, netflows, and perp trades, top holders, token transfers and PnL leaderboards, wallet transaction history, Hyperliquid positions, prediction markets, and more.

Follow [Manually connecting to Nansen MCP](/mcp/connecting.md) for Claude Desktop, Claude Code, Cursor, and other clients.

{% hint style="info" %}
Building an agent that runs from a terminal or scripts? The [Nansen CLI](/cli.md) covers the full API, including alerts and trading, and ships agent skills you can load with `npx skills add nansen-ai/nansen-cli`.
{% endhint %}

### What you can ask

* "Show Solana tokens with Smart Money inflows over the last 24 hours."
* "What are Smart Traders and funds holding on Ethereum right now?"
* "Find Wintermute in Nansen and show its current token holdings."
* "Look up PEPE on Ethereum and summarize its recent price and token flows."
* "Analyze the realized PnL of this wallet over the last 90 days."
* "Who does this address transact with most, and who funded it?"
* "Which prediction markets have the most volume this week?" (full catalog only)

Data coverage varies by chain and tool. See [Data Coverage by Chain](/api/data-coverage.md).

### Using Nansen data elsewhere

Before you republish data returned by these tools, read the [MCP Data Redistribution Guidelines](/mcp/redistribution-guidelines.md).


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.nansen.ai/mcp/overview.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
