Purpose: This is a server based on the Model Context Protocol (MCP) that exposes the capabilities of the enigmapython library to large language models, enabling them to encrypt and decrypt messages using a historically accurate Enigma machine simulator.
Main capabilities: Supports multiple known Enigma models, including M3, M4, I, I_Norway, I_Sondermaschine, K, K_Swiss, D, Z, B_A133, T. Provides the encrypt_message tool, which can dynamically configure rotors (rotor_type, ring_setting, initial_position), reflectors (reflector_type; rotating reflectors can also have ring_setting and initial_position set), and plugboard pairs (plugboard_pairs). The rotor list must be arranged strictly in the order [fastest/rightmost, middle, slowest/leftmost, Greek (M4 only)].
Usage: Requires Python 3.11+. Can be installed from PyPI: pip install enigmapython-mcp, or use uvx enigmapython-mcp. Local MCP clients can use stdio transport: enigmapython-mcp --transport stdio; network mode can use SSE: enigmapython-mcp --transport sse --host 0.0.0.0 --port 8000. Docker is also supported: docker build -t enigmapython-mcp ., then docker run -i enigmapython-mcp (stdio) or docker run -p 8000:8000 enigmapython-mcp --transport sse --host 0.0.0.0 --port 8000. Claude Desktop provides two mcpb one-click installation packages (Docker version and Python version), and claude_desktop_config.json can also be configured manually; OpenCode can be configured in the mcp section of opencode.json.
Use cases: Suitable for scenarios that require an LLM to perform historical Enigma machine encryption and decryption, verify historical ciphertext, or conduct cryptography demonstrations.
Limitations: The materials do not provide information on fees, pricing, performance, or effectiveness; because MCP requires a stateful initialization handshake, the SSE endpoint is relatively complex to test manually with curl, and the official recommendation is to use MCP Inspector for testing.
