The Agent Skills Directory

[PROMPT_INJECTION]: No malicious overrides, bypass attempts, or instructions to ignore safety filters were found. The instructional language is professional and focused on the stated design critique purpose.- [DATA_EXFILTRATION]: No patterns for accessing sensitive files (like .env or .ssh) or exfiltrating data via network requests (curl, fetch, etc.) were detected.- [OBFUSCATION]: No hidden content, Base64-encoded commands, zero-width characters, or homoglyphs were found in the skill metadata or body.- [REMOTE_CODE_EXECUTION]: The skill does not perform any remote code execution or download external scripts. It provides static CSS and HTML code examples for the agent to use as reference.- [EXTERNAL_DOWNLOADS]: No external package installations (npm, pip) or remote resource fetching were detected.- [INDIRECT_PROMPT_INJECTION]: While the skill instructions involve reading external design-context files, it lacks dangerous capabilities (such as file-write or network-access tools) that could be exploited through malicious content in those files.

design-critique