AI agents invoke run_project to trigger actions in Godot Devtool. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Running a Godot project executes all associated GDScript code, assets, and engine operations. While not inherently destructive, the effects depend entirely on what the project does—it could modify files, network state, or system resources.
From the tool's definition Tool name 'run_project' and description 'Run the Godot project and capture output' explicitly indicate execution of external code/project via the Godot engine. This triggers arbitrary game logic, scripts, and side effects determined by the project contents.
Documented attack patterns abuse exactly the kind of access run_project gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Godot Devtool, and nothing reaches the server without passing your rules. This is the rule we recommend for run_project:
{
"version": "1",
"default": "deny",
"tools": {
"run_project": {
"limits": [
{
"counter": "run_project_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_project stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Run the Godot project and capture output. It is categorised as a Execute tool in the Godot Devtool MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Godot Devtool MCP server in PolicyLayer and add a rule for run_project: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Godot Devtool. Nothing to install.
run_project is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the run_project rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for run_project. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
run_project is provided by the Godot Devtool MCP server (wangdiandao/godot-devtool). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 101 Godot Devtool tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
101 Godot Devtool tools catalogued and risk-classified — across an index of 42,500+ MCP servers.