$ loading_
Let AI agents play Z-Machine adventures with session-based MCP access.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "zmachine-mcp" yet — see the docs or source repo.
Use zmachine-mcp to start a new Zork session, keep reading the game text, and choose the next command for the AI agent until it gets the lamp or hits a clear blocker.
A step-by-step sequence of game commands, key story feedback, and the current session or save-state details.
Use zmachine-mcp to load the same interactive fiction scenario, let two AI agents with different strategies play 10 turns each, and compare their choices, score changes, and failure reasons.
A structured comparison showing each command, game response, score progression, and strategic differences.
Use zmachine-mcp over WebSocket to create a persistent session, play part of a Z-Machine game, and record the current state every 5 turns so it can be resumed later.
A resumable session record with checkpoint saves, recent command history, and the current game text.
Automate Zalo personal and official accounts through MCP for AI-driven workflows.
Build, run, and debug real 2D and 3D games via AI conversation.
Connect to the mcp API via MCP to extend AI tool capabilities.
Give MCP agents local-first long-term memory without databases or API keys.
Enable AI agents to play Library of Ruina through game state and UI control.
Expose website JavaScript functions as MCP tools for AI browser interaction.