social-content
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a 'Reverse Engineering' framework that relies on processing and analyzing large quantities of external, untrusted social media data.
- Ingestion points: The agent is instructed to collect and analyze 500-1000+ posts from external creators using tools like Apify or Phantom Buster, as detailed in
references/reverse-engineering.md. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' markers when the agent processes the external scraped text.
- Capability inventory: The agent is tasked with qualitative analysis, pattern extraction, and emotional trigger identification based on this external content.
- Sanitization: The skill lacks instructions for sanitizing or filtering external content to prevent maliciously crafted posts from influencing agent behavior during the analysis phase.
Audit Metadata