Skip to main content
Glama
Erodenn

Godot MCP Runtime

by Erodenn
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript

run_project

Destructive

Launch a Godot project as a child process to enable screenshots, input simulation, UI inspection, and script execution. Verifies MCP bridge readiness and returns the assigned port.

Instructions

Spawn a Godot project as a child process with stdout/stderr captured. Required before take_screenshot, simulate_input, get_ui_elements, run_script, or get_debug_output. Set profiling: true at launch to enable the profiler tools. Use attach_project for one you launched yourself. Verifies MCP bridge readiness before returning success. Returns status with the assigned bridge port. Call stop_project when done. Errors if projectPath is not a Godot project or another session is already active.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
sceneNoScene to run (path relative to project, e.g. "scenes/main.tscn"). Omit to use the project's main scene.
profilingNoAttach Godot's own remote debugger so profile_project, start_profiler and stop_profiler can measure this session. Must be set at launch - a session already running cannot be profiled - and costs a little runtime overhead.
backgroundNoIf true, hides the Godot window off-screen and blocks all physical keyboard and mouse input, while keeping programmatic input (simulate_input, run_script) and screenshots fully active. Useful for automated agent-driven testing where the window should not be visible or interactive.
bridgePortNoTCP port for the MCP bridge. Omit to auto-select a free port (recommended). Delivered to the spawned process via an environment variable, so the bridge script on disk is unaffected by which port this session uses - safe for multiple sessions on the same project.
projectPathYesPath to the Godot project directory

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv3.8.1
    • changedInput schema / properties / bridgePort / description
      Previous value: -"TCP port for the MCP bridge. Omit to auto-select a free port (recommended). The chosen port is baked into the project's `mcp_bridge.gd` at inject time, so the running Godot listens on exactly this port."New value: +"TCP port for the MCP bridge. Omit to auto-select a free port (recommended). Delivered to the spawned process via an environment variable, so the bridge script on disk is unaffected by which port this session uses - safe for multiple sessions on the same project."
  2. Changed1 schema field changedv3.6.0
    • changedInput schema / properties / profiling / description
      Previous value: -"Attach Godot's own remote debugger so profile_project, start_profiler and stop_profiler can measure this session. Must be set at launch — a session already running cannot be profiled — and costs a little runtime overhead."New value: +"Attach Godot's own remote debugger so profile_project, start_profiler and stop_profiler can measure this session. Must be set at launch - a session already running cannot be profiled - and costs a little runtime overhead."
  3. Changed1 schema field changedv3.4.0
    • addedInput schema / properties / profiling
      Added value: +{
+  "description": "Attach Godot's own remote debugger so profile_project, start_profiler and stop_profiler can measure this session. Must be set at launch — a session already running cannot be profiled — and costs a little runtime overhead.",
+  "type": "boolean"
+}
  4. Addedv3.1.1
  5. Removedv3.0.0
  6. Addedv1.0.0

TDQS

A4.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The annotations only declare destructiveHint=true, which is minimal. The description adds substantial behavioral context: it spawns a child process, captures stdout/stderr, verifies MCP bridge readiness before returning, returns a bridge port, and errors if another session is active. It doesn't explicitly state what makes it destructive (e.g., whether it kills existing processes or modifies project files), but it does disclose the key side effects and lifecycle expectations. The description does not contradict the destructiveHint annotation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but well-organized: it front-loads the core action, then lists prerequisites, key options, lifecycle, and error conditions. Every sentence earns its place. It could be slightly more concise by trimming redundancy like 'Required before' and 'Call stop_project when done' being somewhat obvious from the sibling list, but it's efficient and scannable.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 5 parameters, no output schema, and only a destructiveHint annotation, the description covers the essential context: what it does, when to use it, what it returns (status with bridge port), and error conditions. It doesn't describe the exact return format or what 'status' contains, but the absence of an output schema makes that a minor gap. The description is complete enough for an agent to invoke it correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all 5 parameters. The description adds value by explaining the profiling parameter's timing constraint ('Must be set at launch - a session already running cannot be profiled') and the bridgePort behavior ('Delivered to the spawned process via an environment variable... safe for multiple sessions'). This goes beyond the schema's basic descriptions, though the schema already carries most of the semantic weight.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb ('Spawn'), a resource ('Godot project'), and the mechanism ('as a child process with stdout/stderr captured'). It also names the sibling tools it is a prerequisite for, which distinguishes it from attach_project and launch_editor. An agent can tell exactly what this tool does and how it differs from similar tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly says when to use this tool ('Required before take_screenshot, simulate_input, get_ui_elements, run_script, or get_debug_output'), when to use an alternative ('Use attach_project for one you launched yourself'), and when to stop ('Call stop_project when done'). It also states error conditions. This is comprehensive usage guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools