MCP Server¶

By using RISKEN MCP Server , you can connect RISKEN data and security operations to AI models (LLMs) using the Model Context Protocol (MCP) .

Use Cases¶

For example, by using MCP client tools such as Claude Desktop or Cursor , you can perform security operations like:

"(You) Are there any unresolved alerts?" → "(AI) There is 1 alert generated today. This alert is..."
"(You) Search for high-score AWS-related findings" → "(AI) I found 3 items. This finding is..."
"(You) How can this issue be resolved?" → "(AI) The cause is... so it can be resolved by..."
"(You) Archive this finding because it is a false positive" → "(AI) I will archive it."

Setup¶

Obtain an access token from RISKEN.
- For details, refer to Access Token.
Install and start Docker.
Configure Claude Desktop or Cursor with the following settings:

{
  "mcpServers": {
    "risken": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "RISKEN_ACCESS_TOKEN",
        "-e",
        "RISKEN_URL",
        "ghcr.io/ca-risken/risken-mcp-server",
        "stdio"
      ],
      "env": {
        "RISKEN_ACCESS_TOKEN": "your_access_token",
        "RISKEN_URL": "http://your_host"
      }
    }
  }
}

For details on the tools and resources provided by the MCP server, refer to the README.