The Agent Skills Directory

[PROMPT_INJECTION]: No override, bypass, or role-play injection patterns were detected. The instructions are strictly task-oriented and emphasize a controlled, multi-step confirmation workflow.
[DATA_EXFILTRATION]: No hardcoded credentials, sensitive file path access, or unauthorized network operations were identified. The skill operates entirely within the agent's generation context.
[OBFUSCATION]: No obfuscated content, such as Base64, hex encoding, or zero-width characters, was found in any of the provided files.
[REMOTE_CODE_EXECUTION]: No remote scripts, package installations, or dynamic code execution patterns are present.
[COMMAND_EXECUTION]: There are no shell commands or subprocess calls. The skill relies on prompt engineering and standard image generation capabilities.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data (images and text). However, it mitigates risk by employing a structured extraction process (Pass 1 and Pass 2) that maps user input to specific identity and outfit schemas before prompt assembly. No dangerous capabilities are exposed to this input.
[DYNAMIC_CONTEXT_INJECTION]: No use of the dynamic command execution syntax (!command) was found in the SKILL.md file.

portrait-to-turnaround