Ebka AI

A powerful AI-powered assistant for Caido web application security testing, built with Claude AI

Report Bug • Request Feature

Claude Desktop Integration

• Ebka AI
  • Overview
  • Features
    • Request Analysis & Search
    • Replay Session Management
    • Security Findings Management
    • Match/Replace Rules
    • Filters
    • Scopes
    • AI-Powered Intelligence
  • Prerequisites
  • Getting Started
    • Method 1 - Claude Desktop via npx (Recommended)
    • Method 2 - Claude Desktop from Source
  • Codex Integration
    • Codex via npx
    • Codex from Source
  • Replace /absolute/path/to/Ebka-Caido-AI with the path to your local checkout.
  • Installation
    • Prerequisites
    • Install from source:
  • Usage
  • Contributing
  • License

Overview

Ebka AI is an AI-powered assistant that integrates seamlessly with Caido, providing intelligent security testing capabilities through natural language commands and automated workflows. Built with Claude AI, it offers advanced HTTPQL query search, match/replace operations, replay session management, and AI-powered security analysis.

Features

Ebka AI provides 30+ powerful Claude tools for Caido:

Request Analysis & Search

• HTTPQL Query Search: Advanced filtering and analysis using HTTPQL syntax
• Request/Response Viewing: Inspect individual requests and responses by ID
• Custom Request Sending: Send HTTP requests with full control over headers, body, and parameters
• WebSocket Stream Management: Read and analyze WebSocket streams

Replay Session Management

• Replay Collections: Create, list, and manage replay session collections
• Session Operations: Rename sessions, move between collections, and execute automated testing
• Connection Management: Monitor and analyze replay connections and requests

Security Findings Management

• Findings CRUD: Create, read, update, and delete security findings
• Advanced Filtering: List findings with pagination, filtering, and sorting
• Comprehensive Data: Access detailed finding information including request/response bodies

Match/Replace Rules

• Tamper Rule Collections: Organize and manage rule collections
• Rule Management: Create, update, and manage sophisticated find/replace operations
• Advanced Filtering: Search and filter rules by collection and criteria

Filters

• Create and update Filters

Scopes

• Full scope management: CRUD scopes, use created scopes in HTTPQL

AI-Powered Intelligence

• Claude Integration: Leverage Claude AI for intelligent security insights
• Natural Language: Interact with security tools using natural language commands
• Automated Workflows: Streamline security testing with AI-assisted automation

Prerequisites

• Caido web application security testing platform
• Node.js (version 18 or higher) — for MCP server
• For Direct Usage: Claude API key from Anthropic Console

Getting Started

There are multiple ways to interact with the Caido AI Assistant:

Method 1 - Claude Desktop via npx (Recommended)

The easiest way to get started. No need to clone the repository or build anything.

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "caido": {
      "command": "npx",
      "args": ["-y", "ebka-caido-ai"],
      "env": {
        "CAIDO_BASE_URL": "http://localhost:8080"
      }
    }
  }
}

Note: If Caido is running on a different port, update CAIDO_BASE_URL accordingly (e.g., http://localhost:8081).

Authenticate with Caido:

1. Open Claude Desktop and start a conversation
2. When you try to use any Caido tool, Claude will automatically detect that authentication is required
3. Use the authenticate tool to start the OAuth flow
4. Claude will provide a verification URL — click it to authorize in your browser
5. After authorizing, use the check_authentication tool to complete the setup
6. The MCP server will automatically manage your authentication token

That's it! You can now communicate with Caido through Claude.

Method 2 - Claude Desktop from Source

If you prefer to build from source:

1. Clone and build:

   git clone https://github.com/Slonser/Ebka-Caido-AI.git
   cd Ebka-Caido-AI/claude-mcp-server
   npm install
   npm run build
   

2. Add to claude_desktop_config.json:

   {
     "mcpServers": {
       "caido": {
         "command": "node",
         "args": ["/path/to/Ebka-Caido-AI/claude-mcp-server/build/index.js"],
         "env": {
           "CAIDO_BASE_URL": "http://localhost:8080"
         }
       }
     }
   }
   

   Replace /path/to/ with the actual path to your cloned repository.

3. Follow the same authentication steps as described in Method 1.

Codex Integration

Codex can use the same MCP server as Claude Desktop. Add the Caido MCP server to your Codex MCP configuration, then restart Codex so it reloads the tool list.

The examples below assume Caido is listening on http://localhost:8080. If your Caido instance uses another port, update CAIDO_BASE_URL (for example http://127.0.0.1:8088).

Codex via npx

Add this to your Codex config, usually ~/.codex/config.toml:

[mcp_servers.caido]
command = "npx"
args = ["-y", "ebka-caido-ai"]

[mcp_servers.caido.env]
CAIDO_BASE_URL = "http://localhost:8080"

Codex from Source

Use this when developing the MCP server locally or testing unreleased changes:

git clone https://github.com/Slonser/Ebka-Caido-AI.git
cd Ebka-Caido-AI/claude-mcp-server
npm install
npm run build

Then configure Codex:

[mcp_servers.caido]
command = "node"
args = ["/absolute/path/to/Ebka-Caido-AI/claude-mcp-server/build/index.js"]

[mcp_servers.caido.env]
CAIDO_BASE_URL = "http://localhost:8080"

Replace /absolute/path/to/Ebka-Caido-AI with the path to your local checkout.

Installation

Prerequisites

• Caido (latest version)
• Node.js (version 18 or higher)
• npm or pnpm package manager

Install from source:

1. Clone the repository:

   git clone https://github.com/Slonser/Ebka-Caido-AI.git
   cd Ebka-Caido-AI
   

2. Install dependencies:

   pnpm install
   # or if using npm:
   # npm install
   

3. Build the project:

   pnpm build
   # or if using npm:
   # npm run build
   

4. Install in Caido:

   • Open Caido
   • Go to Settings > Plugins
   • Click "Install from file"
   • Select the built plugin file from the appropriate directory

Usage

1. Access Ebka AI:

   • After installation, find "Ebka AI" in your Caido sidebar
   • Click to open the AI assistant interface

2. Configure your settings:

   • Enter your Claude API key for direct usage
   • Configure Claude Desktop integration if using the MCP server
   • Set up your preferred security testing workflows

3. Use AI-powered features:

   • Ask natural language questions about your security testing
   • Use HTTPQL queries to search through requests
   • Create and manage match/replace rules
   • Execute replay sessions and collections
   • Generate security findings with AI assistance

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the GPL-3.0 License - see the LICENSE file for details.