MCP server

The Model Context Protocol (MCP) server provides a standardized interface that allows any compatible AI model or agent to access Takara's AI research paper summaries and search functionality in a simple and secure way.

Connect to our MCP server natively in Claude, Cursor, and other clients. Takara's MCP server follows the remote MCP spec with HTTP transport, so the server is centrally hosted and managed.

Setup Instructions

General

Our MCP server uses HTTP transport. The endpoint is:

https://tldr.takara.ai/api/mcp

Claude

Navigate to Settings in Claude and click on "Connectors" in the sidebar
Scroll to the bottom and click "Add custom connector"
Enter the following URL and click "Add":
```
https://tldr.takara.ai/api/mcp
```
Complete any authentication prompts if required
Configure tool permissions in the connector settings as needed

Claude Code

Run the following command in Terminal:

claude mcp add --transport http takara-tldr https://tldr.takara.ai/api/mcp

Then run /mcp once you've opened a Claude Code session to activate the server.

Codex

The setup steps for the MCP server are the same regardless of whether you use the IDE Extension or the CLI since the configuration is shared.

Configuration via CLI:

Run the following command in Terminal:

codex mcp add takara-tldr --url https://tldr.takara.ai/api/mcp

Configuration through config file:

Open the ~/.codex/config.toml file in your preferred editor. Add the following:

[features]
experimental_use_rmcp_client = true

[mcp_servers.takara-tldr]
url = "https://tldr.takara.ai/api/mcp"

Note: If this is the first time you are using an MCP in Codex you will need to enable the rmcp feature for this to work. The experimental_use_rmcp_client = true setting is included above.

Cursor

To add the MCP to Cursor, click the button below for one-click installation, or search for Takara TLDR from Cursor's MCP tools page.

Alternatively, add the following to your Cursor settings:

{
  "mcpServers": {
    "takara-tldr": {
      "url": "https://tldr.takara.ai/api/mcp",
      "transport": "http"
    }
  }
}

Visual Studio Code

CTRL/CMD P and search for MCP: Add Server.
Select HTTP
Enter the following configuration, and hit enter:
```
https://tldr.takara.ai/api/mcp
```
Enter the name Takara TLDR and hit enter.
Activate the server using MCP: List Servers and selecting Takara TLDR, and selecting Start Server.

Or add directly to your settings:

{
  "mcpServers": {
    "takara-tldr": {
      "url": "https://tldr.takara.ai/api/mcp",
      "transport": "http"
    }
  }
}

v0 by Vercel

To add the MCP to v0, you can install from the connections page.

Windsurf

CTRL/CMD , to open Windsurf settings.
Under Scroll to Cascade → MCP servers
Select Add Server → Add custom server

Add the following:

{
  "mcpServers": {
    "takara-tldr": {
      "url": "https://tldr.takara.ai/api/mcp",
      "transport": "http"
    }
  }
}

Zed

CMD , to open Zed settings.

Add the following:

{
  "context_servers": {
    "takara-tldr": {
      "source": "custom",
      "url": "https://tldr.takara.ai/api/mcp",
      "transport": "http"
    }
  }
}

Others

Hundreds of other tools now support MCP servers. You can configure them to use Takara's MCP server with the following settings:

URL: https://tldr.takara.ai/api/mcp
Transport: http

{
  "url": "https://tldr.takara.ai/api/mcp",
  "transport": "http"
}

Available Tools

get_latest_summary

Get the latest daily AI research paper summary. Returns structured JSON data with title, date, headline, content, and papers. All paper links point to tldr.takara.ai pages where you can read full summaries and explore related research.

get_paper

Get detailed information about a specific research paper by its ID (e.g., ArXiv ID like "1706.03762" or "2405.15953"). Returns full paper data including title, abstract, authors, publication date, and links. The primary 'url' and 'tldrUrl' fields point to tldr.takara.ai where you can read the full paper summary and explore related research.

search_papers

Search for research papers by query using semantic search with embedding-based reranking. Returns a list of matching papers with titles, summaries, metadata, and URLs pointing to tldr.takara.ai pages where you can read full summaries and explore related research. Supports an optional limit parameter (default: 20, max: 100).

Stay in the loop

Get tldr.takara.ai to Your Email, Everyday.

Content is sourced from third-party publications.