The Agent Skills Directory

[PROMPT_INJECTION]: No malicious prompt injection patterns were found. The 'Critical Rules' section contains domain-specific architectural guidelines for Unity development rather than attempts to bypass AI safety filters.
[DATA_EXFILTRATION]: There are no network operations, hardcoded credentials, or accesses to sensitive file paths. The skill focuses entirely on local Unity project architecture.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns were detected. The skill does not download external scripts or execute arbitrary commands from remote sources.
[OBFUSCATION]: The content is presented in plain text and standard C# code. No hidden characters, Base64 encoding, or homoglyph attacks were identified.
[PRIVILEGE_ESCALATION]: The provided shell script scripts/run.sh does not use sudo or other commands to escalate privileges. It is a standard wrapper for executing the agent's Python runner.
[DYNAMIC_CONTEXT_INJECTION]: No use of dynamic command execution patterns (e.g., backticks with exclamation marks) was found in the SKILL.md file.

unity-architect