Kagi MCP Server

What is Kagi MCP Server about?

Enables AI agents to perform real‑time web searches and generate summaries of URLs (including videos) through a standardized MCP interface. It connects Kagi's search and summarizer APIs with Claude Desktop, Claude Code, and any other client that supports MCP.

How to use Kagi MCP Server?

1. Install the UV tool (required to run the Python server).
2. Clone the repository and set up a virtual environment with uv venv.
3. Activate the environment and install dependencies via uv sync.
4. Configure the server in Claude Desktop or Claude Code by adding an MCP entry that runs uvx kagimcp and supplies KAGI_API_KEY (and optionally KAGI_SUMMARIZER_ENGINE).
5. Issue queries that need a tool, e.g., a search request or a video‑summarization request, and the server will handle the API calls and return results.

Key features of Kagi MCP Server

• Web search using Kagi's beta search API.
• Content summarization for articles and videos with selectable engines (default cecil).
• Easy integration with Claude Desktop, Claude Code, and any MCP‑compatible client.
• Customizable logging via FASTMCP_LOG_LEVEL.
• Environment‑driven configuration for API keys and summarizer engine.

Use cases of Kagi MCP Server

• Augmenting AI assistants with up‑to‑date web information.
• Summarizing long videos or articles on‑the‑fly for downstream processing.
• Providing reliable search results within private or corporate AI workflows.
• Replacing default web‑search tools in Claude with a curated Kagi source.

FAQ from the Kagi MCP Server

Q: Do I need a Kagi account? A: Yes, access to the search API is in closed beta; request an invite via support@kagi.com.

Q: Which summarizer engines are supported? A: Set KAGI_SUMMARIZER_ENGINE to any engine listed in Kagi’s API docs (e.g., cecil, daphne). The default is cecil.

Q: How do I disable Claude's built‑in web search to avoid conflicts? A: Add a deny entry for WebSearch in Claude’s settings.json permissions.

Q: Can I run the server on Windows? A: Yes. Install UV using PowerShell, create the virtual environment, then activate .venv\Scripts\activate before running uv sync and uvx kagimcp.

Q: Where can I see debugging information? A: Run the inspector with npx @modelcontextprotocol/inspector uvx kagimcp and open the web UI at http://localhost:5173.

Kagi MCP server

Setup Intructions

Before anything, unless you are just using non-search tools, ensure you have access to the search API. It is currently in closed beta and available upon request. Please reach out to support@kagi.com for an invite.

Install uv first.

MacOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Installing via Smithery

Alternatively, you can install Kagi for Claude Desktop via Smithery:

npx -y @smithery/cli install kagimcp --client claude

Setup with Claude

Claude Desktop

// claude_desktop_config.json
// Can find location through:
// Hamburger Menu -> File -> Settings -> Developer -> Edit Config
{
  "mcpServers": {
    "kagi": {
      "command": "uvx",
      "args": ["kagimcp"],
      "env": {
        "KAGI_API_KEY": "YOUR_API_KEY_HERE",
        "KAGI_SUMMARIZER_ENGINE": "YOUR_ENGINE_CHOICE_HERE" // Defaults to "cecil" engine if env var not present
      }
    }
  }
}

Claude Code

Add the Kagi mcp server with the following command (setting summarizer engine optional):

claude mcp add kagi -e KAGI_API_KEY="YOUR_API_KEY_HERE" KAGI_SUMMARIZER_ENGINE="YOUR_ENGINE_CHOICE_HERE" -- uvx kagimcp

Now claude code can use the Kagi mcp server. However, claude code comes with its own web search functionality by default, which may conflict with Kagi. You can disable claude's web search functionality with the following in your claude code settings file (~/.claude/settings.json):

{
  "permissions": {
    "deny": [
      "WebSearch"
    ]
  }
}

Pose query that requires use of a tool

e.g. "Who was time's 2024 person of the year?" for search, or "summarize this video: https://www.youtube.com/watch?v=jNQXAC9IVRw" for summarizer.

Debugging

Run:

npx @modelcontextprotocol/inspector uvx kagimcp

Local/Dev Setup Instructions

Clone repo

git clone https://github.com/kagisearch/kagimcp.git

Install dependencies

Install uv first.

MacOS/Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Then install MCP server dependencies:

cd kagimcp

# Create virtual environment and activate it
uv venv

source .venv/bin/activate # MacOS/Linux
# OR
.venv/Scripts/activate # Windows

# Install dependencies
uv sync

Setup with Claude Desktop

Using MCP CLI SDK

# `pip install mcp[cli]` if you haven't
mcp install /ABSOLUTE/PATH/TO/PARENT/FOLDER/kagimcp/src/kagimcp/server.py -v "KAGI_API_KEY=API_KEY_HERE"

Manually

# claude_desktop_config.json
# Can find location through:
# Hamburger Menu -> File -> Settings -> Developer -> Edit Config
{
  "mcpServers": {
    "kagi": {
      "command": "uv",
      "args": [
        "--directory",
        "/ABSOLUTE/PATH/TO/PARENT/FOLDER/kagimcp",
        "run",
        "kagimcp"
      ],
      "env": {
        "KAGI_API_KEY": "YOUR_API_KEY_HERE",
        "KAGI_SUMMARIZER_ENGINE": "YOUR_ENGINE_CHOICE_HERE" // Defaults to "cecil" engine if env var not present
      }
    }
  }
}

Pose query that requires use of a tool

e.g. "Who was time's 2024 person of the year?" for search, or "summarize this video: https://www.youtube.com/watch?v=jNQXAC9IVRw" for summarizer.

Debugging

Run:

# If mcp cli installed (`pip install mcp[cli]`)
mcp dev /ABSOLUTE/PATH/TO/PARENT/FOLDER/kagimcp/src/kagimcp/server.py

# If not
npx @modelcontextprotocol/inspector \
      uv \
      --directory /ABSOLUTE/PATH/TO/PARENT/FOLDER/kagimcp \
      run \
      kagimcp

Then access MCP Inspector at http://localhost:5173. You may need to add your Kagi API key in the environment variables in the inspector under KAGI_API_KEY.

Advanced Configuration

• Level of logging is adjustable through the FASTMCP_LOG_LEVEL environment variable (e.g. FASTMCP_LOG_LEVEL="ERROR")
  • Relevant issue: https://github.com/kagisearch/kagimcp/issues/4
• Summarizer engine can be customized using the KAGI_SUMMARIZER_ENGINE environment variable (e.g. KAGI_SUMMARIZER_ENGINE="daphne")
  • Learn about the different summarization engines here