The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override system instructions, bypass safety filters, or perform jailbreak attacks were detected across any of the 16 files.- [DATA_EXFILTRATION]: No commands for accessing sensitive local files (e.g., .ssh, .aws) or exfiltrating data to external domains are present. The skill focuses entirely on game development logic.- [REMOTE_CODE_EXECUTION]: The skill does not contain scripts, shell commands, or patterns that download and execute code from remote servers.- [COMMAND_EXECUTION]: No shell commands or subprocess executions are defined. The instructions are purely conversational and instructional for the agent's behavior.- [CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, passwords, or private keys were found in the instructions or knowledge modules.- [EXTERNAL_DOWNLOADS]: The skill mentions various industry-standard Unity services and frameworks (such as Unity Netcode, Mirror, and Addressables) as topics of expertise, but it does not initiate any automated downloads or external connections.- [OBFUSCATION]: Analysis of the markdown content revealed no hidden characters, Base64-encoded commands, homoglyph substitutions, or other obfuscation techniques.

unity-developer