Dradis MCP

A FastMCP server implementation for interacting with Dradis Pro, a collaborative platform for information security teams. This MCP allows you to manage projects and vulnerabilities in Dradis directly through your MCP-enabled tools.

Features

• Project Management
  • Create new projects
  • Get project details
  • Set active project for operations
• Vulnerability Management
  • Create vulnerabilities
  • Update existing vulnerabilities
  • Get vulnerability details
  • Get all vulnerability details
  • List vulnerabilities with pagination support
• Content Block Management
  • Get all content blocks in current project
  • Update a content block
• Document Property Management
  • Get all document properties
  • Create or update a document property
• Robust error handling and input validation
• Consistent JSON response formatting

Prerequisites

• Node.js (v14 or higher)
• npm
• A Dradis Pro instance
• Dradis API token

Installation

npm install

Configuration

Environment Variables

Create a .env file in the root directory with the following variables:

DRADIS_URL=<url>
DRADIS_API_TOKEN=<token>
DRADIS_DEFAULT_TEAM_ID=<teamId>
DRADIS_DEFAULT_TEMPLATE_ID=<templateId>
DRADIS_VULNERABILITY_PARAMETERS=<commaSeparatedParams> // Title,Description,Solution,etc. 

MCP Config File

To add Dradis MCP to your MCP configuration, add the following to your config file to mcp.json if you're using 5ire:

{
  "servers": [
    {
      "key": "Dradis",
      "command": "npx",
      "description": "A Model Context Protocol server that provides integration with Dradis note-taking platform",
      "args": ["-y", "dradis-mcp"],
      "env": {
        "DRADIS_URL": "<url:string:Your Dradis instance URL>",
        "DRADIS_API_TOKEN": "<token:string:Your Dradis API token>",
        "DRADIS_DEFAULT_TEAM_ID": "<number:string:Default team ID for project creation>",
        "DRADIS_DEFAULT_TEMPLATE_ID": "<number:string:Default template ID for project creation>"
      },
      "isActive": false
    }
  ]
}

Available Tools

Project Management

• setProject: Set the current Dradis project

  { projectId: number }

• getProjectDetails: Get details of the current project

  // No parameters required

• createProject: Create a new Dradis project

  {
    name: string;
    team_id?: number;  // Optional if DRADIS_DEFAULT_TEAM_ID is set
    report_template_properties_id?: number;  // Optional if DRADIS_DEFAULT_TEMPLATE_ID is set
    author_ids?: number[];
    template?: string;  // Optional if DRADIS_DEFAULT_TEMPLATE is set
  }

Vulnerability Management

• createVulnerability: Create a new vulnerability

  {
    text: string; // Content with #[ ]# field syntax
  }

• getVulnerabilities: List vulnerabilities (25 per page)

  {
    page?: number; // Optional page number
  }

• getVulnerability: Get specific vulnerability

  {
    vulnerabilityId: number;
  }

• updateVulnerability: Update existing vulnerability

  {
    issueId: number;
    parameters: {
      text: string;
    }
  }

Content Block Management

• getContentBlocks: Get all content blocks in current project

  // No parameters required

• updateContentBlock: Update a content block

  {
    blockId: number;
    contentBlock: {
      content: string;
    }
  }

Document Property Management

• getDocumentProperties: Get all document properties

  // No parameters required

• upsertDocumentProperty: Create or update a document property

  {
    propertyName: string;
    value: string;
  }

Development

To run the server in development mode with the CLI:

npm run dev

To inspect the server using MCP Inspector:

npm run inspect

Building

To build the TypeScript code:

npm run build

Running in Production

To run the built server:

npm start

Project Structure

• /src - Source code
  • api.ts - Dradis API client implementation
  • config.ts - Configuration loading and validation
  • server.ts - MCP server implementation
  • types.ts - TypeScript type definitions
• /dist - Compiled JavaScript (generated after build)

Error Handling

The MCP provides detailed error messages for:

• Missing or invalid configuration
• API request failures
• Invalid input parameters
• Missing project ID
• Network errors

Response Format

All tool responses are formatted as JSON with consistent structure:

• Success responses include relevant data and optional success messages
• Error responses include detailed error messages and context
• List endpoints include pagination metadata when applicable