agentsclimarketplace

Roblox creator store security audit

Skill AshExplained/roblox-skills/roblox-creator-store-security-audit

34 Claude Code AI skills for Roblox game development: ideas, Luau architecture, gameplay systems, UX, economy, monetization, security, QA, publishing, and live ops.

Install
npx -y skills add AshExplained/roblox-skills --skill roblox-creator-store-security-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Audit Roblox Creator Store, Toolbox, inventory, group, universe, package, model, UI, gameplay, plugin-like, and other third-party imported assets for malicious or risky scripts. Use whenever Claude imports, inserts, reviews, or trusts Creator Store assets, especially assets with Scripts, LocalScripts, ModuleScripts, hidden descendants, require calls, HTTP, DataStore, remote creation, obfuscated code, or suspicious names.

SKILL.md

4.3 KB, as published. Nobody here has run it

Roblox Creator Store Security Audit

Goal

Treat every third-party Creator Store or Toolbox import as untrusted until inspected. Roblox documentation identifies third-party assets as a security risk because they can contain malicious scripts/backdoors, and Creator Store assets can include scripts. Audit before enabling, shipping, or building on imported behavior.

Source Baseline

When current policy or platform behavior matters, verify official Roblox docs first:

  • https://create.roblox.com/docs/scripting/security/third-party-vulnerabilities
  • https://create.roblox.com/docs/scripting/capabilities
  • https://create.roblox.com/docs/production/creator-store

Use the audit workflow even if Roblox adds platform protections. Platform safeguards reduce risk; they do not replace reading and understanding imported scripts.

Import Rule

For third-party model/package/gameplay/UI assets:

  1. Insert into a quarantine folder or isolated location.
  2. Do not enable HTTP requests or other security-sensitive settings because an imported asset prompts for it.
  3. Disable or remove imported scripts unless the game explicitly needs them and they pass review.
  4. Prefer visual-only extraction: keep meshes, parts, textures, sounds, particles, and UI art; remove unknown behavior.
  5. Record asset name, asset ID, creator/source, inserted path, and audit result.

MCP Audit Workflow

  1. Confirm Studio connection with list_roblox_studios, set_active_studio if needed, and get_studio_state.
  2. Locate the imported asset with search_game_tree.
  3. Inspect the imported root with inspect_instance.
  4. Search code with script_grep for suspicious patterns:
    • require
    • getfenv
    • setfenv
    • loadstring
    • HttpService
    • GetObjects
    • InsertService
    • MarketplaceService
    • DataStore
    • MessagingService
    • TeleportService
    • RemoteEvent
    • RemoteFunction
    • rbxassetid
    • asset
    • string.reverse
    • string.char
    • byte
    • spawn
    • coroutine
    • while true
  5. Use script_search and script_read on every imported Script, LocalScript, and ModuleScript.
  6. Inspect descendants for scripts hidden under unexpected objects, disabled scripts, deceptive names, or deeply nested containers.
  7. If scripts are not needed, remove or disable them with multi_edit or targeted execute_luau.
  8. Playtest in isolation with start_stop_play, then check get_console_output.

Red Flags

Treat as high risk:

  • Obfuscated or unusually complex code.
  • Code that fetches or executes external modules by asset ID.
  • Prompts to enable HttpService, paste command-bar code, or install another asset/plugin.
  • Unknown remotes, admin panels, backdoor-looking commands, or hidden GUIs.
  • Scripts named like AntiLag, MainModule, Loader, Handler, Update, Version, HDAdmin, Free, Fix, or names unrelated to the asset purpose.
  • DataStore writes, purchase handling, teleporting, moderation/admin commands, or player messaging inside a decorative asset.
  • Code that reparents itself, clones itself, destroys audit-visible objects, or spreads through services.

Verdicts

Return one of:

  • Safe to keep visual content only: scripts removed or unnecessary.
  • Safe with reviewed scripts: scripts are simple, necessary, and explained.
  • Quarantine: suspicious but not proven malicious; do not ship.
  • Reject/remove: malicious, obfuscated, backdoor-like, or unjustified sensitive behavior.

Output

Report:

  • Asset name, ID/source, creator when known, and inserted path.
  • Script inventory and suspicious-pattern findings.
  • Actions taken: disabled, removed, retained, or unchanged.
  • Security-sensitive settings checked.
  • Final verdict and follow-up tests.

Next

Once an asset is cleared, bring it in through roblox-asset-pipeline and check its runtime cost with roblox-performance-optimization.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.