Skip to main content
Glama
Tann2019

Godot Forge

by Tann2019
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Local

Project files

files

Manage Godot project resources: read, write, search, edit scripts, scenes, and assets. Writes sync the editor, reload scripts, and return compile errors.

Instructions

Read, write, search and manage files inside the Godot project (res://). Writes keep the editor in sync: scripts are reloaded and their compile errors returned, scenes/resources are refreshed, assets reimported. Reading an image returns it as a picture.

Actions:

  • list: {path='res://', recursive=true, pattern?='*.gd', include_addons?, limit?} list files with their resource type.

  • search: {query, regex?, case_sensitive?, glob?, path?, limit?} grep text files (gd, tscn, tres, gdshader, cfg, json, md...).

  • read: {path, offset?, limit?} read a text file (line range optional) or view an image.

  • write: {path, content, overwrite?=true} create/replace a file. .gd files come back with 'diagnostics'.

  • edit: {path, edits: [{old, new, replace_all?}]} exact string replacements (old must match exactly once, tabs matter). Prefer this over rewriting whole files.

  • move: {from, to, update_references?=true} move/rename, updating res:// references in scenes/scripts.

  • copy: {from, to}

  • delete: {path} move a file/folder to the OS trash.

  • mkdir: {path} create a folder (and parents).

  • exists: {path} does a file/folder exist, and its resource type.

  • dependencies: {path} what a resource depends on and what uses it.

  • rescan: {reimport?: [paths]} rescan the filesystem / force reimport (after changing files outside the editor).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
toNoDestination path.
fromNoSource path.
pathNores:// path (bare relative paths are treated as res://).
editsNoExact replacements.
queryNoSearch text or regex.
actionYesWhat to do. See the tool description for each action's parameters.
contentNoFull file content.
patternNoFilename glob.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.7/5.0
Behavior5/5

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

Beyond the annotations, it discloses real behavioral traits: writes keep the editor in sync (scripts reloaded, compile errors returned, scenes/resources refreshed, assets reimported), reads of images return pictures, and delete moves to OS trash rather than permanently removing. The trash behavior and editor-sync side effects are exactly the context an agent needs and are not derivable from destructiveHint=false alone.

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 purpose is front-loaded, then a compact one-line-per-action list that is easy to scan. It is long, but the length is justified by 12 actions; only minor redundancy (parameters repeated in the prose) keeps it from a 5.

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

Completeness5/5

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

For a 12-action multiplexed tool with no output schema, the description covers every action, its parameters, defaults, and side effects, and even previews return values ('resource type', 'diagnostics'). Nothing essential for correct invocation is missing.

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

Parameters5/5

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

The input schema only exposes 8 top-level properties and marks additionalProperties as unconstrained, so most action-specific parameters (recursive, pattern default '*.gd', include_addons, limit, regex, case_sensitive, glob, offset, overwrite default true, update_references, reimport) exist only in the description. The per-action parameter lists with defaults and the warning that edit's 'old must match exactly once, tabs matter' add substantial semantics over the schema.

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 opening sentence names a specific verb set (read, write, search, manage) and a precise resource (files inside the Godot project at res://), then enumerates all 12 actions with their key parameters. This clearly differentiates it from siblings like script, resource, scene, and assets, which cover narrower domains.

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

Usage Guidelines4/5

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

Usage context is given for several actions: 'Prefer this over rewriting whole files' for edit, 'after changing files outside the editor' for rescan, and the note that .gd writes return diagnostics. There is no explicit routing guidance versus sibling tools (e.g. when to use files+read instead of the script tool), so it stops short of a 5.

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