Godot Genre Metroidvania logo

Godot Genre Metroidvania

Community
thedivergentai
godot-genre-metroidvania

Expert blueprint for Metroidvanias including ability-gated exploration (locks/keys), interconnected world design (backtracking with shortcuts), persistent state tracking (collectibles, boss defeats), room transitions (seamless loading), map systems (grid-based revelation), and ability versatility (combat + traversal). Use for exploration platformers or action-adventure games. Trigger keywords: metroidvania, ability_gating, interconnected_world, backtracking, map_system, persistent_state, room_transition, soft_locks.

Overview

Publisherthedivergentai
RepositoryGD-Agentic-Skills
Skill namegodot-genre-metroidvania
Stars
727
Forks
43
Bundled files
20
LicenseLGPL-3.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 20 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by thedivergentai on GitHub. Read the source before you install it.

Installation

Install the Godot Genre Metroidvania AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/thedivergentai/GD-Agentic-Skills.git /tmp/GD-Agentic-Skills
mkdir -p .claude/skills
cp -r /tmp/GD-Agentic-Skills/skills/godot-genre-metroidvania .claude/skills/godot-genre-metroidvania
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Godot Genre Metroidvania in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Godot Genre Metroidvania on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Godot Genre Metroidvania is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

NEVER Do (Expert Anti-Patterns)

World Design & Exploration

  • NEVER allow "Soft-Locks" where a player is trapped; if they enter via a one-way path ("valve"), they MUST be able to leave using current abilities. Always design fail-safe escape routes.
  • NEVER create empty dead ends; if a player backtracks to a remote area, they MUST be rewarded with a collectible, lore, or currency. Empty rooms are design failures.
  • NEVER make backtracking purely repetitive; as the player gains movement (Dash/Teleport), traversal through old areas MUST become faster. Open shortcuts to bypass long, early routes.
  • NEVER hide the critical path without "crumbs"; use distinct Landmarks, unique lighting, or environmental storytelling to build the player's mental map.
  • NEVER design abilities that serve only one purpose; strictly implement dual-use traversal and combat functionality (e.g., a "Dash" that crosses gaps and dodges attacks).

Persistence & Mapping

  • NEVER forget to save persistent room state; if a player opens a chest or defeats a boss, that state MUST remain saved when they leave and return.
  • NEVER load interconnected rooms synchronously via load(); strictly use ResourceLoader.load_threaded_request() for seamless transitions.
  • NEVER track global progression within localized room scripts; strictly use Autoload Singletons for global ability flags and world state.
  • NEVER use floating-point types for grid coordinates (minimaps/fog); strictly use Vector2i to prevent precision jitter.
  • NEVER manipulate the SceneTree directly from a background loading thread; strictly use call_deferred().

Physics & Controls

  • NEVER calculate jump arcs or dashes inside _process(); strictly use _physics_process() to prevent stutter.
  • NEVER multiply CharacterBody2D velocity by delta before move_and_slide(); the engine handles this internally.
  • NEVER poll is_action_just_pressed() inside _physics_process() for buffering; strictly capture events in _unhandled_input().
  • NEVER use standard strings for high-frequency ability checks; strictly use StringName (&"dashing") for pointer-speed comparisons.
  • NEVER iterate through every node to broadcast updates; strictly use SceneTree.call_group() for efficient mass communication.
  • NEVER delete active room/player nodes via free(); strictly use queue_free() to avoid segmentation faults.

🛠 Expert Components (scripts/)

Golden-path order (MANDATORY reads)

  1. Persistence / flags — metroid_game_state.gd + persistent_progression_system.gd
  2. Ability definitions / gates — ability_unlock_resource.gd + progression_gate_manager.gd
  3. Room stream / switch — background_room_streamer.gd + safe_scene_switcher.gd
  4. Map fog — minimap_fog_manager.gd (orchestrator) with minimap_fog.gd / minimap_fog_revealer.gd

Original Expert Patterns

Modular Components


Core Loop

  1. Exploration → blocked by a lock
  2. Discovery → key ability / boss
  3. Acquisition → new traversal/combat tool
  4. Backtracking → shortcuts + dual-use abilities
  5. Progression → new biome opens

Skill Chain

PhaseSkillsPurpose
1. Charactergodot-characterbody-2d, godot-state-machine-advancedTight movement + ability states
2. Worldgodot-tilemap-mastery, godot-scene-managementRooms, biomes, threaded transitions
3. Systemsgodot-save-load-systems, godot-ability-systemPersist gates/collectibles; unlock keys
4. UIgodot-ui-containers, godot-inventory-systemMap / inventory / HUD
5. Balancegodot-monte-carlo-balancerSoft-lock risk, backtrack length

Architecture (script-first — no inline recipes)

1. Game State & Persistence

MANDATORY: metroid_game_state.gd + persistent_progression_system.gd. Rooms never own global ability flags. Room metadata uses resource_local_to_scene so instanced rooms do not share collectible state.

2. Room Transitions & Fast Travel

MANDATORY: background_room_streamer.gd + safe_scene_switcher.gd.

Fast travel must match NEVER (threaded load + deferred swap) — never ResourceLoader.load() / sync change_scene:

gdscript
class_name FastTravelSystem extends Node

var _pending_path: String = ""
var _spawn_id: StringName = &""

func travel_to_room(scene_path: String, spawn_id: StringName) -> void:
    _pending_path = scene_path
    _spawn_id = spawn_id
    var err := ResourceLoader.load_threaded_request(scene_path)
    if err != OK:
        push_error("Fast travel request failed: %s" % scene_path)
        return
    set_process(true)

func _process(_delta: float) -> void:
    var status := ResourceLoader.load_threaded_get_status(_pending_path)
    if status == ResourceLoader.THREAD_LOAD_IN_PROGRESS:
        return
    set_process(false)
    if status != ResourceLoader.THREAD_LOAD_LOADED:
        push_error("Fast travel load failed: %s" % _pending_path)
        return
    var packed := ResourceLoader.load_threaded_get(_pending_path) as PackedScene
    # SceneTree work must be deferred — never from a worker thread
    call_deferred("_swap_room", packed, _spawn_id)

func _swap_room(packed: PackedScene, spawn_id: StringName) -> void:
    GlobalState.target_spawn_id = spawn_id
    get_tree().change_scene_to_packed(packed)

3. Ability Gating

MANDATORY: ability_unlock_resource.gd + progression_gate_manager.gd + ability_state_machine.gd. Gates query StringName abilities from the Autoload — do not hardcode ability strings in room scripts.

4. Map / Fog

MANDATORY: minimap_fog_manager.gd. Use Vector2i cells only.


Design Principles (from Dreamnoid)

  • Ability Versatility — traversal + combat dual use
  • Practice Rooms — teach before punish
  • Landmarks — mental map without explicit markers
  • Item micro-stories — lore without cutscene walls

Common Pitfalls

  1. Softlocks on one-way valves — always design an escape with current abilities
  2. Backtracking tedium — shortcuts + faster movement after unlocks
  3. Empty dead ends — every remote path needs a reward
  4. Sync room loads — violates NEVER; use threaded request + deferred swap

Expert knowledge (on demand)

LLM-ignorance rule: If a general agent would not know it before reading, load the reference — never delete expert deltas.

Reference

Progressive disclosure: open Official Documentation links only when researching a specific API; load Related Skills when routing to a peer domain — do not preload the whole lattice.

Official Documentation

  • Using CharacterBody2Dmove_and_slide, floor detection, and velocity rules for coyote/buffer jumps and dash traversal.
  • Idle and Physics Processing — keep jump arcs, dashes, and wall slides in _physics_process; capture buffers in input callbacks.
  • InputEvent_unhandled_input jump/ability buffering so presses are not lost between physics ticks.
  • Background loadingResourceLoader.load_threaded_request for adjacent-room preload without hitch spikes.
  • Change scenes manually — deferred room swaps, spawn door IDs, and safe queue_free of the outgoing room.
  • Saving games — persist abilities, opened gates, collectibles, and visited map cells across sessions.
  • Singletons (Autoload) — global progression/game-state ownership so rooms never keep conflicting ability flags.
  • Resources — ability unlock definitions and room metadata as .tres data with safe duplication.
  • Using TilemapsTileMapLayer + Vector2i cells for minimap fog revelation and grid room tracking.
  • Camera2D — room limit_* bounds and tweened limit handoffs for seamless camera room transitions.
  • Using Area2D — doors, save stations, and hazard triggers via body_entered without hard scene coupling.
  • Groupscall_group for save-station heal/respawn broadcasts instead of walking the whole tree.

Related Skills

Prerequisites
  • godot-project-foundations — autoloads, scene layout, and project settings before stacking room streaming and global progression.
  • godot-characterbody-2d — tight platformer locomotion is the substrate under ability-gated traversal (dash, wall slide, double jump).
  • godot-tilemap-mastery — layered TileMap/TileMapLayer authorship for gameplay collision, landmarks, and minimap fog grids.
  • godot-autoload-architecture — singleton ownership patterns for ability flags and world persistence that rooms must not duplicate.
Complements
  • godot-scene-management — threaded load queues and deferred room switches that keep interconnected maps hitch-free.
  • godot-save-load-systems — durable schemas for collectibles, boss flags, and visited cells across long exploration sessions.
  • godot-camera-systems — room limits, RemoteTransform follow, and transition polish beyond basic Camera2D bounds.
  • godot-ability-system — unlockable traversal/combat abilities that gates and state machines query as the “keys.”
  • godot-state-machine-advanced — hierarchical player states for dash/wall-slide/double-jump without nested if/elif sprawl.
  • godot-2d-physics — layers, Area2D doors/hazards, and direct space queries for wall detection and soft-lock-safe valves.
  • godot-inventory-system — collectible/key item tracking that feeds map rewards and ability acquisition UI.
Downstream / consumers
  • godot-monte-carlo-balancer — simulate ability unlock order, backtrack length, and soft-lock risk once gates and reward density are tunable.
  • godot-genre-platformer — pure movement-feel patterns that Metroidvania traversal builds on when stripping ability gating.
  • godot-signal-architecture — ability_unlocked / gate_opened / map_revealed buses so HUD and rooms observe progression without owning it.
Master
  • godot-master — library router and mirrored module entry; use when discovering peer skills or syncing shared script mirrors after Domain Skill edits.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Godot Genre Metroidvania AI skill do?

Expert blueprint for Metroidvanias including ability-gated exploration (locks/keys), interconnected world design (backtracking with shortcuts), persistent state tracking (collectibles, boss defeats), room transitions (seamless loading), map systems (grid-based revelation), and ability versatility (combat + traversal). Use for exploration platformers or action-adventure games. Trigger keywords: metroidvania, ability_gating, interconnected_world, backtracking, map_system, persistent_state, room_transition, soft_locks.

Why use Godot Genre Metroidvania on TypingMind?

Because you install it once and use it with any model. Godot Genre Metroidvania is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Godot Genre Metroidvania in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/thedivergentai/GD-Agentic-Skills/tree/main/skills/godot-genre-metroidvania. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Godot Genre Metroidvania?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Godot Genre Metroidvania?

As many as you like. As long as a model supports skills, you can use Godot Genre Metroidvania with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Godot Genre Metroidvania AI skill free?

Yes. It is published on GitHub by thedivergentai under the LGPL-3.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇