Hashcat

You can run Hashcat MCP Server to integrate intelligent hash cracking directly with Claude Desktop, enabling hash identification, multiple attack modes, batch processing, and password analysis from your conversations. This server is designed to be configured locally and interacted with via an MCP client, giving you secure, automated control over your cracking workflows.

How to use

Install and run the Hashcat MCP Server to enable hash cracking capabilities within your MCP client. You can crack single hashes or process multiple hashes in batches, automatically identify hash types, and run auto attack strategies to optimize your efforts. Use the provided client commands and sample prompts to initiate identifications, attacks, or analyses, then monitor progress and review results within your conversation flow.

How to install

pip install -r requirements.txt

Install Dependencies

• Run the package manager to install required Python dependencies listed in requirements.txt.

Configure Claude Desktop Windows: %APPDATA%\Claude\claude_desktop_config.json Mac: ~/Library/Application Support/Claude/claude_desktop_config.json

{
    "mcpServers": {
        "Hashcat-MCP": {
            "command": "python",
            "args": [
                "C:\\path\\to\\your\\hashcat_mcp_server.py"
            ],
            "env": {
                "HASHCAT_PATH": "C:\\path\\to\\hashcat\\hashcat.exe",
                "HASHCAT_SAFE_DIRS": "C:\\wordlists"
            }
        }
    }
}

Important: Update the paths to match your system:

• args[0] - Path to your hashcat_mcp_server.py file
• HASHCAT_PATH - Path to your hashcat executable
• HASHCAT_SAFE_DIRS - Comma-separated list of safe directories for wordlists

3. Restart Claude Desktop Close and reopen Claude Desktop to load the MCP server.

## Additional configuration and usage notes

Required Environment Variables
- `HASHCAT_PATH` - Path to hashcat executable (required)

Optional Environment Variables
- `HASHCAT_SAFE_DIRS` - Safe directories for wordlists (recommended)
- `HASHCAT_LOG_LEVEL` - Logging level: DEBUG, INFO, WARNING, ERROR (default: DEBUG)

Usage Examples
- Basic Hash Identification: `Can you identify this hash: 5d41402abc4b2a76b9719d911017c592`

## Security and notes

This server is intended for authorized security testing and educational purposes only. Ensure you have explicit permission before testing any hashes or systems.

## Troubleshooting

If Claude Desktop cannot load the MCP server after editing the config, ensure the paths are correct and that the Python environment has access to the hashcat executable. Check that the hashcat\_mcp\_server.py script is present at the specified path and that required dependencies are installed.

## Available tools

### smart\_identify\_hash

Identify hash types with confidence scores to guide attack strategies.

### crack\_hash

Crack a single hash using various attack modes such as dictionary, brute-force, hybrid, or rule-based approaches.

### crack\_multiple\_hashes

Process and crack a batch of hashes in one go to improve throughput.

### auto\_attack\_strategy

Automatically select and run effective attack sequences tailored to the hash type.

### benchmark\_hashcat

Benchmark hashcat performance on your hardware to assess capability.

### get\_gpu\_status

Monitor real-time GPU usage and health during cracking.

### analyze\_cracked\_passwords

Analyze cracked passwords to extract security insights and patterns.

### generate\_smart\_masks

Create intelligent brute-force masks for efficient password discovery.

### estimate\_crack\_time

Estimate the time required for attacks based on configured settings.