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.
npx -y skills add AshExplained/roblox-skills --skill roblox-creator-store-security-auditAssembled 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-vulnerabilitieshttps://create.roblox.com/docs/scripting/capabilitieshttps://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:
- Insert into a quarantine folder or isolated location.
- Do not enable HTTP requests or other security-sensitive settings because an imported asset prompts for it.
- Disable or remove imported scripts unless the game explicitly needs them and they pass review.
- Prefer visual-only extraction: keep meshes, parts, textures, sounds, particles, and UI art; remove unknown behavior.
- Record asset name, asset ID, creator/source, inserted path, and audit result.
MCP Audit Workflow
- Confirm Studio connection with
list_roblox_studios,set_active_studioif needed, andget_studio_state. - Locate the imported asset with
search_game_tree. - Inspect the imported root with
inspect_instance. - Search code with
script_grepfor suspicious patterns:requiregetfenvsetfenvloadstringHttpServiceGetObjectsInsertServiceMarketplaceServiceDataStoreMessagingServiceTeleportServiceRemoteEventRemoteFunctionrbxassetidassetstring.reversestring.charbytespawncoroutinewhile true
- Use
script_searchandscript_readon every importedScript,LocalScript, andModuleScript. - Inspect descendants for scripts hidden under unexpected objects, disabled scripts, deceptive names, or deeply nested containers.
- If scripts are not needed, remove or disable them with
multi_editor targetedexecute_luau. - Playtest in isolation with
start_stop_play, then checkget_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.