social-content

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The 'Reverse Engineering Viral Content' framework in 'references/reverse-engineering.md' instructs the agent to ingest and analyze large datasets of external social media posts, creating a surface for indirect prompt injection.
  • Ingestion points: The 'SCRAPE' section in 'references/reverse-engineering.md' specifies collecting 'Post text/content' and other metadata from hundreds of external social media posts.
  • Boundary markers: The skill does not provide instructions to treat this external data as untrusted or to use delimiters to prevent command confusion.
  • Capability inventory: The agent uses the data to derive patterns and generate content, which could be influenced by malicious prompts in the source data.
  • Sanitization: No sanitization or validation steps are defined for the external data before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 01:23 PM