> Design save/load for game state — choosing what to serialize, file formats, save slots, atomic crash-safe writes, schema versioning and migration, and autosave. Engine-neutral. Use when the user mentions save system, save/load, game state persistence, save slots, autosave, save file corruption, or migrating old saves to a new version.
npx skills add https://github.com/gamedev-skills/awesome-gamedev-agent-skills --skill save-systems
A save file is a serialized snapshot of game state that survives restarts.
The hard parts aren't writing bytes — they're choosing *what* to save, writing it
so a crash mid-save can't corrupt it, and reading *old* saves after you ship a
patch. Get those three right and the rest is plumbing.
positions — across sessions and game updates.
migration).
When *not* to use: for Roblox cloud persistence specifics, use
roblox-datastores. For the data model the save serializes (resources/SOs), use
godot-resources / unity-scriptableobjects. For Godot's FileAccess/
ResourceSaver and user:// paths, defer to the Godot engine skill while
applying the patterns here.
unlocked flags), not engine objects or scene nodes. You will reconstruct
objects from data on load — never serialize live node references.
version integer. This isthe single most important field for a game you intend to patch.
format for size/speed or mild tamper-resistance. Start with JSON.
real file. A crash leaves either the old save or the new one — never a
half-written one.
instantiate. Keep a backup of the last good save and fall back on parse error.
separate slot so it can't clobber a manual save.
state matches. Test loading a save from the previous version.
# Build a dictionary of pure data. Each savable object reports its own state.
func capture_state() -> Dictionary:
return {
"version": SAVE_VERSION, # ALWAYS stamp the schema version
"player": { "hp": player.hp, "pos": [player.position.x, player.position.y] },
"inventory": player.inventory.to_array(), # ids + counts, not Item nodes
"flags": world.flags, # e.g. {"met_guard": true}
"seed": world.seed, # regenerate procedural content
}
# On load, RECONSTRUCT objects from the data — do not expect live references back.
func apply_state(data: Dictionary) -> void:
player.hp = data["player"]["hp"]
player.position = Vector2(data["player"]["pos"][0], data["player"]["pos"][1])
player.inventory.from_array(data["inventory"])
world.flags = data["flags"]
# RIGHT: write to a temp file, then atomically rename over the target.
func save_atomic(path: String, data: Dictionary) -> void:
var tmp := path + ".tmp"
var f := FileAccess.open(tmp, FileAccess.WRITE)
f.store_string(JSON.stringify(data))
f.flush() # ensure bytes hit disk
f.close()
DirAccess.rename_absolute(tmp, path) # replaces the target; atomic on POSIX
# WRONG: opening `path` directly and writing in place — a crash mid-write leaves a
# truncated, unloadable save and destroys the player's progress.
Rename-over-target is atomic on POSIX (same volume); on Windows a replace-by-rename
isn't guaranteed atomic, so keep the previous file as path + ".bak" before the
rename — that backup is what actually guarantees you can recover from a bad write.
SAVE_VERSION = 3
def load_save(raw_bytes):
data = parse(raw_bytes) # JSON/binary -> dict
v = data.get("version", 0)
if v > SAVE_VERSION:
raise NewerSaveError(v) # save is from a newer build; refuse
while v < SAVE_VERSION: # apply migrations in order, v -> v+1
data = MIGRATIONS[v](data)
v += 1
data["version"] = v
validate(data) # check required keys / ranges
return data
# Each migration is a pure function from one version's shape to the next.
def migrate_1_to_2(d):
d["flags"] = {k: True for k in d.pop("completed_quests", [])} # list -> set-map
return d
MIGRATIONS = {1: migrate_1_to_2, 2: migrate_2_to_3}
const SLOT_PATH := "user://save_%d.json" # manual slots 0..N
const AUTOSAVE_PATH := "user://autosave.json" # separate file: never clobbers a slot
var _autosave_cooldown := 0.0
func autosave_if_due(dt: float) -> void:
_autosave_cooldown -= dt
if _autosave_cooldown <= 0.0:
save_atomic(AUTOSAVE_PATH, capture_state())
_autosave_cooldown = 60.0 # throttle: at most once a minute
# Trigger an immediate autosave on checkpoints/level transitions, not mid-combat.
renaming a node breaks every old save. Save data, rebuild objects on load.
guessing game. Stamp version from version 1.
rename; keep a .bak.
cloud-synced stale. Validate on load and fall back to backup on failure.
decimal separators in some locales. Use a locale-invariant serializer.
inconsistent state. Use a dedicated autosave slot and save on safe boundaries.
player-controlled; never treat it as authoritative for online state. For cloud,
handle the device's data limits and conflicts (roblox-datastores).
references/versioning-and-migration.md — schema evolution strategies, themigration chain, backups/rollback, format trade-offs (JSON vs binary), and a
load-time validation checklist.
roblox-datastores — cloud persistence, request limits, session locking.godot-resources, unity-scriptableobjects — the data model you serialize.procedural-gen — store the seed to regenerate worlds instead of saving them.rpg, survival-crafting, visual-novel — genres that compose this skill.Integration with protocols.io API for managing scientific protocols. This skill should be used when working with protocols.io to search, create, update, or publish protocols; manage protocol steps and materials; handle discussions and comments; organize workspaces; upload and manage files; or integrate protocols.io functionality into workflows. Applicable for protocol discovery, collaborative protocol development, experiment tracking, lab protocol management, and scientific documentation.
Analyzes job descriptions and generates tailored resumes that highlight relevant experience, skills, and achievements to maximize interview chances
Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", or "generate an Excalidraw file". Supports flowcharts, relationship diagrams, mind maps, and system architecture diagrams. Outputs .excalidraw JSON files that can be opened directly in Excalidraw.
Build and distribute Expo development clients locally or via TestFlight
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.
Benchling R&D platform integration. Access registry (DNA, proteins), inventory, ELN entries, workflows via API, build Benchling Apps, query Data Warehouse, for lab data management automation.
Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use bioservices.
Take gamedev-skills/save-systems from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.