July 2026
Unity Code MCP Server v0.5 Release Notes
The 0.5 release removes the UniTask requirement and makes gameplay input, async tool execution, and Unity package compatibility more reliable.
Available Now
Unity Code MCP Server
A direct bridge between your AI agent and Unity Editor.
Created for long running AI agents that need to do real work in your Unity project, the open-source Unity Code MCP Server gives AI agents first-class access to the Unity Editor.
It turns Unity automation into a practical loop: inspect the live project, act directly through Unity Editor APIs or simulated player input, then validate the result against real runtime state.
It allows agents to understand the context, like scene, assets or settings, make focused changes, run gameplay checks, and keep iterating with observable feedback.
Inspect the live Unity Editor project
Give your agent the context it needs before it touches anything. Unity Code MCP Server lets MCP clients inspect the active scene, project assets, component data, console output, and Play Mode state.
That means the agent can understand how this specific project is put together instead of guessing.
- Discover scene objects, prefabs, assets, components, and runtime values.
- Read Unity Console output and project settings as part of the same investigation.
- Build a grounded plan from real Editor state before executing changes.
Act directly inside the Unity Editor
Agents can run C# in the Editor process to create objects, wire scenes, update assets, prepare test state, and automate the repetitive project work that usually slows iteration.
For runtime workflows, the same agent can enter Play Mode and drive real input actions.
- Author and update scenes, prefabs, ScriptableObjects, import settings, components and other assets or settings.
- Use Unity Editor APIs through generated C# scripts.
- Drive gameplay and UI through input actions when validation needs player-like behavior.
Verify with runtime feedback
Close the loop by checking what actually happened. The agent can run Edit Mode and Play Mode tests, enter Play Mode, emulate player input, read logs, capture and analyze screenshots and inspect live runtime data after every action.
This turns AI-assisted development into a repeatable inspect / act / verify cycle: find the state, act on it, measure the result, and continue until the scenario passes or the failure is clear.
- Validate player movement, UI flows, collisions, game state, and scene startup behavior.
- Combine runtime inspection, console logs, screenshots, and test results.
- Shorten feedback loops with automated playtesting.
Built-in tools
Built-in tools connect the agent to the active Unity project. The agent can execute C# with full Editor API access, read Console output and project state, run Edit Mode and Play Mode tests, control Play Mode, simulate configured Input System actions, and capture Game View screenshots. These operations support project inspection, implementation, reproduction, and verification.
- execute_csharp_script_in_unity_editor
- Execute generated C# scripts with full Unity Editor API access and modify scenes, assets, configuration and more.
- read_unity_console_logs
- Read live Unity Console output from the current project.
- run_unity_tests
- Run EditMode or PlayMode tests and bring the results back into the agent workflow.
- enter_play_mode, exit_play_mode
- Move the editor into Play Mode for gameplay automation or testing.
- play_unity_game
- Automate gameplay with input actions so agents can validate in-game behavior.
- get_unity_game_view_window_screenshot
- Capture the current Game View screenshot.
- get_unity_info
- Return information about the current Unity Editor project and MCP server settings.
Latest release
Requirements
- Unity 6000.3 or higher. Older versions may work but are not officially supported.
- Unity Input System for gameplay automation; version 1.19.0 is declared by the package.
- Newtonsoft Json 3.2.2, declared and resolved through Unity Package Manager.
- uv for STDIO transport workflows.
Security risks
Unity Code MCP Server executes AI-generated C# with Unity Editor privileges. Use it at your own risk. To the fullest extent permitted by law, Signal Loop disclaims liability for any changes, damage, or data loss resulting from its use. You are responsible for securing your environment.