The Agent Skills Directory

[NO_CODE]: The skill consists entirely of Markdown-based instructions and a JSON evaluation file. There are no scripts, binaries, or automated command-line operations present in the package.
[PROMPT_INJECTION]: No malicious injection patterns or attempts to override safety protocols were detected. The instructions focus purely on the intended brand audit methodology.
[DATA_EXFILTRATION]: The skill does not contain any network-enabled commands (e.g., curl, wget) or patterns that suggest exfiltration of sensitive data.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests external data (brand guidelines and website content) for auditing, it is a low-risk analytical task. It provides a structured framework for the agent to follow, minimizing the risk of accidental instruction following from untrusted data.

brand-audit