The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to override system instructions or bypass safety filters were found. The instructional text is focused on color theory and academic standards.
[DATA_EXFILTRATION]: No network operations, hardcoded credentials, or access to sensitive local file paths (like .ssh or .aws) were identified.
[EXTERNAL_DOWNLOADS]: All external links point to well-known, trusted academic and design tools (e.g., colorbrewer2.org, Adobe Color, WebAIM). These are used for legitimate research and accessibility verification.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs such as 'venue' and 'diagram type'. While it lacks explicit boundary markers or sanitization instructions for this input, the risk is negligible as the skill's logic is purely advisory and does not involve executing the input in a high-privilege environment.
[COMMAND_EXECUTION]: Although 'bash' is listed in the allowed-tools metadata, the skill does not contain any scripts, commands, or instructions that would invoke the shell for malicious purposes.
[OBFUSCATION]: No obfuscated content, such as base64-encoded strings, zero-width characters, or homoglyphs, was detected. The hexadecimal color codes used are standard CSS values.

Academic Figure Color Expert