The Agent Skills Directory

[PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety guidelines or override agent behavior. The content is strictly focused on Roblox development.
[DATA_EXFILTRATION]: No network-related commands (like HttpService) or sensitive data access patterns were detected. The skill operates entirely within the context of the Roblox engine environment.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or other credentials were found in the scripts or documentation.
[REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts. All code is locally defined within the provided files.
[COMMAND_EXECUTION]: The Lua scripts use standard Roblox API calls. No shell command execution or unsafe system-level operations are present.
[OBFUSCATION]: The code and documentation are written in clear, human-readable text. No hidden characters, Base64 encoding of executable content, or homoglyph-based attacks were identified.
[PRIVILEGE_ESCALATION]: The skill does not include any commands or instructions designed to gain unauthorized permissions or elevate access levels within the development environment.
[PERSISTENCE]: No mechanisms for maintaining persistent access or creating backdoor entries were detected.
[DYNAMIC_EXECUTION]: The scripts do not use unsafe dynamic execution functions like loadstring. It uses standard module requiring for local file structures.
[INDIRECT_PROMPT_INJECTION]: The skill does not ingest untrusted external data that could be used to manipulate agent prompts. It focuses on game world state and positions.

roblox-pathfinding-and-npcs