- Home
- Skills
- Game Development
- Godot Safe Change — Guarded MCP Workflow
Godot Safe Change — Guarded MCP Workflow
Guide AI agents to preview, confirm, apply, verify, and safely roll back bounded Godot project changes.
Free
Works with the AI Agents you already use
About this skill
Godot Safe Change gives MCP-capable AI agents a guarded workflow for changing Godot projects. It starts with the live editor context, previews bounded changes, requires explicit confirmation and revision/lease checks, applies through Godot's EditorPlugin or guarded file routes, verifies the resulting state, and only rolls back while the original history and revision still match.
Requirements
This skill provides agent instructions. It does not include the MCP server or Godot EditorPlugin. Install and connect godot-safe-change-mcp, enable the Godot Safe Change Bridge plugin in a Godot 4.x editor, and use an MCP-capable agent client. Node.js 22.22.2 or newer is required.
Setup and supported operations: https://github.com/LJH-snow/godot-safe-change-mcp#readme
What you get
- A preview, explicit confirmation, apply, and verification workflow.
- Guidance for supported scene, script, resource, input-action, and project-setting changes.
- Revision and lease checks that reject stale or conflicting changes.
- Guarded rollback that refuses to overwrite later user changes.
Boundaries
Only operations exposed by the connected server are supported. No arbitrary Godot RPC, arbitrary script execution, or unrestricted filesystem writes. Editor operations require a connected bridge. This is a free, MIT-licensed skill and project.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Requires
Permissions
Reads project context and performs bounded project-local writes through the separately installed MCP server and editor bridge. Requires explicit confirmation, revision and lease checks. The bridge uses local loopback networking. Terminal access is for setup and read-only git inspection; arbitrary shell or script execution is not a fallback for mutations.