MCP Integration

Connect Prismor directly to your AI assistant using the Model Context Protocol (MCP). Get vulnerability scans and security context without leaving your editor.

What is MCP?

The Model Context Protocol (MCP) allows AI assistants like Cursor, Windsurf, and Claude to interact with external tools.

By adding the Prismor MCP server, your AI assistant gains the ability to:

Scan your repo dependencies for vulnerabilities
Check for secrets violations

Server Details

Server URL

https://mcp.prismor.dev/mcp

Authentication

Authorization: Bearer YOUR_API_KEY

How to Use

Once configured, you can simply ask your AI assistant to scan your project.

Example Prompt

"Scan and fix vulnerabilities in my repository.."

Cursor

Step 1: Open MCP Settings

Go to Cursor Settings → General → MCP

Step 2: Add New MCP Server

Click Add new global MCP server and select "HTTP"

Alternatively, you can directly edit ~/.cursor/mcp.json

Step 3: Add Configuration

json

{
  "mcpServers": {
    "prismor": {
      "url": "https://mcp.prismor.dev/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Note: Replace YOUR_API_KEY with your actual Prismor API key from the dashboard.

Windsurf

Step 1: Locate Configuration File

Open your Windsurf MCP configuration file:

~/.codeium/windsurf/mcp_config.json

Step 2: Add Prismor Configuration

Add the following to your configuration file:

json

{
  "mcpServers": {
    "prismor": {
      "serverUrl": "https://mcp.prismor.dev/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Note: After saving the configuration, restart Windsurf for changes to take effect.

VS Code

Step 1: Install MCP Extension

First, ensure you have a compatible MCP extension installed in VS Code (such as GitHub Copilot with MCP support).

Step 2: Open Settings

Open VS Code Settings (Cmd/Ctrl + ,) and search for "MCP" to find MCP server configuration.

Step 3: Configure MCP Server

Add the following to your VS Code MCP configuration file. See VS Code MCP docs for more info.

json

{
  "mcp": {
    "servers": {
      "prismor": {
        "type": "http",
        "url": "https://mcp.prismor.dev/mcp",
        "headers": {
          "Authorization": "Bearer YOUR_API_KEY"
        }
      }
    }
  }
}

Important: MCP support in VS Code may require specific extensions or VS Code Insiders version. Check the latest documentation for compatibility.

Cline

Step 1: Access MCP Servers

Open Cline and click the hamburger menu icon (☰) to enter the MCP Servers section.

Step 2: Manual Configuration

Alternatively, you can directly edit the MCP servers configuration:

json

{
  "mcpServers": {
    "prismor": {
      "url": "https://mcp.prismor.dev/mcp",
      "type": "streamableHttp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Claude Code

Using Command Line

Run this command to add Prismor MCP server. See Claude Code MCP docs for more info.

bash

claude mcp add --transport http prismor https://mcp.prismor.dev/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"

Note: Replace YOUR_API_KEY with your actual Prismor API key.

Gemini CLI

Step 1: Locate Settings File

Open the Gemini CLI settings file:

~/.gemini/settings.json

Step 2: Add Configuration

Add the following to the mcpServers object:

json

{
  "mcpServers": {
    "prismor": {
      "httpUrl": "https://mcp.prismor.dev/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY",
        "Accept": "application/json, text/event-stream"
      }
    }
  }
}

Note: If the mcpServers object doesn't exist, create it.

OpenAI Codex

Configuration

Add the following configuration to your OpenAI Codex MCP server settings:

toml

[mcp_servers.prismor]
args = ["-y", "@prismor/mcp", "--api-key", "YOUR_API_KEY"]
command = "npx"

Note: See OpenAI Codex documentation for more information.

Get Your API Key