dwarf_fortress_mcp
Active·★ 24·MIT·Updated 2026-10-21
Dwarf Fortress MCP is a robust control plane enabling autonomous agents to manage Dwarf Fortress as a long-lived civilization. It provides semantic, transactional, and replayable operations for reliable game state monitoring and agent-driven actions.
#Autonomous Agents#Control Plane#Semantic Transactions#Dwarf Fortress Automation#Multi-Agent Systems#Game AI#Replayable Operations#State Management
01
Features
01Compact, cursor-based semantic observations instead of full dumps.
02Multi-version canonical world with provenance, stable identities, and append-only history.
03Semantic intents compiled into immutable, capability-checked action DAGs with various witnesses.
04Prepare/revalidate/commit semantics with idempotency, phantom protection, and deterministic merge.
05Multi-agent ownership leases and MVCC validation for concurrent operation.
02
Compatibility
Linux
OS
Verified via docs
Windows
OS
Verified via docs
macOS
OS (via compatibility layer)
Verified via docs
Rust
Language/Runtime
Verified via docs
Dwarf Fortress
Game
Verified via docs
DFHack
Modding Interface
Verified via docs
03
Quick start
1
$ cargo test --locked --workspace
2
$ cargo run --locked -p dwarf-fortress-mcp -- serve
04
Use cases
↳Long-horizon civilization management in Dwarf Fortress by AI agents.
↳Automating complex tasks like resource management, defense, and construction within the game.
↳Enabling multi-agent collaboration for different aspects of fortress stewardship.
↳Developing and testing robust AI agents for simulated environments.
05
Alternatives
dos-kernel★ 20
Catch your AI agents when they lie about what they shipped — verifies claims against git instead of believing the agent.
Context-Engineering-for-Multi-Agent-Systems★ 290
Save thousands of lines of code by building universal, domain-agnostic Multi-Agent Systems (MAS) through high-level semantic orchestration. This repository provides a production-ready blueprint for the Agentic Era, allowing you to replace rigid, hard-coded workflows with a dynamic transparent Context Engine that provides 100% transparency.
AI-company★ 373
Multi-agent team operating system for Claude Code. 108 MCP tools, 40+ agent templates, 10 lifecycle hooks, 7 pipeline workflows. Persistent teams, structured meetings, task wall, real-time React dashboard. No LangChain/AutoGen — pure CC native integration.
Comments
Log in to leave a comment
No comments yet. Be the first!