unity-mcp and unity-api-mcp
These are complements: one provides direct Unity Editor automation via MCP protocol, while the other provides efficient API reference lookups, and both would typically be used together by an AI assistant to fully interact with and understand a Unity project.
About unity-mcp
CoplayDev/unity-mcp
Unity MCP acts as a bridge, allowing AI assistants (like Claude, Cursor) to interact directly with your Unity Editor via a local MCP (Model Context Protocol) Client. Give your LLM tools to manage assets, control scenes, edit scripts, and automate tasks within Unity.
Implements a comprehensive Python-based MCP server using HTTP transport that operates locally within the Unity Editor, exposing 40+ tools across physics, animation, graphics, builds, and scene management alongside introspection resources like `unity_reflect` for live API inspection. The architecture supports multi-instance routing, async long-running operations with polling, batch execution for performance optimization, and integrates with popular LLM clients (Claude, Cursor, VS Code Copilot, Windsurf) through configurable connection protocols. Available as a Git-based package in the Unity Package Manager, Asset Store, and OpenUPM with Python 3.10+ and the `uv` package manager as runtime dependencies.
About unity-api-mcp
Codeturion/unity-api-mcp
Instant, accurate Unity API lookups instead of expensive source file reads, saving your agent tokens, context, and hallucinations
Implements an MCP server with version-specific SQLite databases (~18–24 MB each) for Unity 2022, 2023, and 6, parsed from official XML documentation and C# source. Provides five tools—search, method signatures, namespace resolution, class references, and deprecation checks—all returning results in <15ms with BM25-ranked search and tuned column weights. Integrates with Claude Code, Cursor, Windsurf, and any MCP client; auto-detects Unity version from `ProjectSettings/ProjectVersion.txt` or explicit environment variables.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work