playbook-paid-social-advertising
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional text and templates for generating marketing documents. No command-line execution or shell scripting is present.
- [DATA_EXPOSURE]: While the skill requests client business data (e.g., WhatsApp numbers and URLs), this is standard input for the stated purpose of building an advertising plan. The skill does not automate the collection or transmission of this data; it merely instructs the agent on what to ask the user.
- [REMOTE_CODE_EXECUTION]: There are no external downloads or remote scripts referenced. All references point to local skill files (e.g.,
caption-writer/SKILL.md), which is a standard pattern for modular instructions. - [OBFUSCATION]: A thorough scan for Base64, zero-width characters, and homoglyphs returned no findings. The text is plain markdown.
- [PROMPT_INJECTION]: The instructions follow a professional tone and do not attempt to bypass safety filters or override system-level agent behavior.
Audit Metadata