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

# SEA-LION MCP Server

SEA-LION Model Context Protocol (MCP) Server provides a simple way to access SEA-LION’s multilingual models and Southeast Asian language tools from MCP-compatible clients.

The Model Context Protocol (MCP) is an open standard for connecting AI models to external tools, data sources, and services. It defines a common interface so that clients (such as IDEs, chat assistants, or AI agents) can discover and use capabilities provided by servers without custom integrations for each tool. Powered by the SEA-LION API, the MCP server exposes capabilities including translation and localisation, and safety classification.

Usage of the SEA-LION API is subject to our [Terms of Use](https://sea-lion.ai/terms-of-use/) and [Privacy Policy](https://sea-lion.ai/privacy-policy/)

## **Connect via Claude Desktop**

Claude Desktop handles the OAuth flow for you automatically. It's the most straightforward way to connect.

1. Open Claude Desktop and go to **Settings → Connectors → Customize**
2. Click Add Custom Connector and select Web
3. Enter a name (e.g. SEA-LION MCP) and the URL: <https://mcp.sea-lion.ai>
4. Click Add and Connect

<figure><img src="/files/jbqrvbZk2nq27UQcPbtl" alt="Claude Desktop Add custom connector dialog with the SEA-LION MCP name and URL filled in" width="70%"><figcaption></figcaption></figure>

5. Click on **Allow Access** and sign in with your Google account

<figure><img src="/files/e8z14DvqjVPRf3PPStSl" alt="SEA-LION Application Access Request screen with the Allow Access button" width="70%"><figcaption></figcaption></figure>

## Connect via Claude Code CLI

To add the MCP server to Claude Code CLI:

```bash
claude mcp add --transport http sealion https://mcp.sea-lion.ai/
```

You'll then be prompted to sign in: your browser opens, and you authorise access with your Google account.

Verify the connection with:

```bash
claude mcp list
```

You should see `sealion` listed as ✓ Connected.

## Using Claude

Once integrated, SEA-LION MCP works invisibly in the background. Users can simply talk to the main LLM such as Claude naturally.

**Example Prompt:**

Can you translate this article into Malay for our Singapore audience?

<figure><img src="/files/NkB1uVMDI385tg5xQW2r" alt="Translation prompt with Claude requesting permission to use the SEA-LION Translate localize tool" width="100%"><figcaption></figcaption></figure>

<figure><img src="/files/3cHEVhze3PvWc6heB3BW" alt="The article translated into Malay and localized for a Singaporean audience" width="100%"><figcaption></figcaption></figure>

Claude automatically calls SEA-LION's translation engine, localising the content for the right tone and context.

Beyond translation, SEA-LION can also detect Southeast Asian language variants and moderate content with cultural context specific to the region.

## Rate Limits

Limits help us mitigate misuse and manage API capacity and help ensure that everyone has fair access to the API.

SEA-LION MCP usage frequency will be subject to rate limits applied on requests per minute (RPM).

As of 04 Jun 2026, our rate limits is set to **10 requests per minute per user**.

If you have any questions or want to speak about getting a rate limit increase, reach out to <sealion@aisingapore.org>.


---

# 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://docs.sea-lion.ai/guides/tool_calling/mcp.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.