Save Load logo

Save Load

Organization
SummerEngine
save-load

Save and load game state: user:// paths, JSON progress vs ConfigFile settings, versioned saves, autosave.

Overview

PublisherSummerEngine
Repositorysummer
Skill namesave-load
Stars
91
Forks
8
Bundled files
1
LicenseMIT
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.

  • 1 bundled files

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

  • Open source

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

Installation

Install the Save Load 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.
    https://github.com/SummerEngine/summer/tree/main/library/skills/save-load
  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/SummerEngine/summer.git /tmp/summer
mkdir -p .claude/skills
cp -r /tmp/summer/library/skills/save-load .claude/skills/save-load
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Save Load 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 Save Load 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 Save Load 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.

Save / Load

Trigger phrases: "save progress", "settings menu should remember", "high scores", "checkpoints", "save slots", "JSON or ConfigFile?", "saves broke after the update".

Save what the player would be angry to lose, and nothing else. Persist DATA (numbers, ids, flags), never live nodes or scene references.

Where saves go

Always user:// - a writable per-user dir (res:// is read-only in an export).

  • user://save_0.json - progress
  • user://settings.cfg - options

Never write gameplay saves to res://; it fails in the shipped game.

JSON vs ConfigFile

Use JSON whenUse ConfigFile when
Nested/structured game state (inventory, quests, world)Flat key/value settings
One document you load wholeSectioned options (audio, video, controls)
You want a schema + version fieldYou want human-editable ini-style

Default to JSON for game progress, ConfigFile for settings.

Save schema (version it from day one)

A top-level version field is the difference between "old saves still load after an update" and "everyone loses their progress on patch day".

gdscript
const SAVE_PATH := "user://save_0.json"
const SAVE_VERSION := 1

func build_save() -> Dictionary:
	return {
		"version": SAVE_VERSION,
		"level": GameManager.current_level,
		"player": {
			"hp": GameManager.player_health,
			"gold": GameManager.player_gold,
			"pos": var_to_str(player.global_position),  # Vector serialized as string
		},
		"inventory": GameManager.inventory,   # array of item ids, not item nodes
		"flags": GameManager.story_flags,
	}

Serialize Vectors/Colors with var_to_str() and restore with str_to_var() - JSON has no native Vector2/3.

Write and read

gdscript
func save_game() -> void:
	var f := FileAccess.open(SAVE_PATH, FileAccess.WRITE)
	if f == null:
		push_error("save failed: %s" % FileAccess.get_open_error())
		return
	f.store_string(JSON.stringify(build_save(), "\t"))   # tab-indented, readable
	f.close()

func load_game() -> bool:
	if not FileAccess.file_exists(SAVE_PATH):
		return false
	var f := FileAccess.open(SAVE_PATH, FileAccess.READ)
	var data: Variant = JSON.parse_string(f.get_as_text())
	f.close()
	if typeof(data) != TYPE_DICTIONARY:
		return false                       # corrupt/partial file
	data = _migrate(data)                  # bring old versions up to current
	_apply_save(data)
	return true

Migration (do not skip)

When the schema changes, bump SAVE_VERSION and upgrade old saves instead of rejecting them.

gdscript
func _migrate(data: Dictionary) -> Dictionary:
	var v: int = data.get("version", 0)
	if v < 1:
		data["flags"] = data.get("flags", {})   # field added in v1
		data["version"] = 1
	return data

Settings with ConfigFile

gdscript
func save_settings() -> void:
	var cfg := ConfigFile.new()
	cfg.set_value("audio", "master", master_volume)
	cfg.set_value("video", "fullscreen", is_fullscreen)
	cfg.save("user://settings.cfg")

func load_settings() -> void:
	var cfg := ConfigFile.new()
	if cfg.load("user://settings.cfg") != OK:
		return                              # first run, use defaults
	master_volume = cfg.get_value("audio", "master", 1.0)

Always pass a default to get_value so a missing key never crashes.

Autosave hooks

  • Save on meaningful checkpoints (level end, area transition, shop close), not every frame. Disk writes stutter.
  • Save on quit via _notification:
gdscript
func _notification(what: int) -> void:
	if what == NOTIFICATION_WM_CLOSE_REQUEST:
		save_game()
		get_tree().quit()

Set get_tree().set_auto_accept_quit(false) so you control the quit.

Traps

  • Storing a node/PackedScene/callable in the save = unserializable or huge. Store ids and rebuild.
  • Writing to res:// - read-only in exports; silently fails or errors.
  • No version field - the first schema change wipes every player's save.
  • JSON.parse_string returns null on bad input; type-check before using it.
  • Floats round-trip fine; ints may load back as floats from JSON - cast on read (int(data["gold"])).

Applying it in Summer Engine

  • The save code is a script, usually an autoload (res://scripts/autoloads/save_manager.gd): create it with summer_write_file(path=..., content=..., create_only=true), then register it with summer_project_setting(key="autoload/SaveManager", value="*res://scripts/autoloads/save_manager.gd") (the leading * makes it a singleton). Check it with summer_get_script_errors.
  • user:// is the per-user data dir, outside the project, so do not expect summer_read_file (a project-file reader) to show a save file. Prove a round trip from inside the game instead: a RunVerification probe (see playtesting-a-feature) that calls save_game(), changes a value, calls load_game(), and report()s the restored value.

From a shell

No Summer MCP tools in this session? Run each one as a command; when summer --version is missing or below 4.0, use npx -y summer-engine@latest in place of summer. Fields are flags in kebab-case (--scene-path res://main.tscn) or one --input '<json>', and summer ops describe <op> prints the exact input. The summer-cli skill has the rest.

In this skillCommand
summer_get_diagnosticssummer engine get-diagnostics
summer_get_script_errorssummer engine get-script-errors
summer_project_settingsummer engine project-setting
summer_read_filesummer engine read-file
summer_replace_textsummer engine replace-text
summer_write_filesummer engine write-file

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 Save Load AI skill do?

Save and load game state: user:// paths, JSON progress vs ConfigFile settings, versioned saves, autosave.

Why use Save Load on TypingMind?

Because you install it once and use it with any model. Save Load 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 Save Load in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SummerEngine/summer/tree/main/library/skills/save-load. 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 Save Load?

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 Save Load?

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

Is the Save Load AI skill free?

Yes. It is published on GitHub by SummerEngine under the MIT 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 👇