social-content

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements a workflow for 'Reverse Engineering Viral Content' that relies on processing untrusted data from external platforms.\n
  • Ingestion points: The 'SCRAPE' section in SKILL.md instructs the agent to gather 500-1000+ posts from social media platforms (LinkedIn, Twitter/X) using scraping tools or manual collection.\n
  • Boundary markers: No boundary markers, delimiters, or explicit instructions to ignore embedded commands within the ingested content are provided.\n
  • Capability inventory: The 'ANALYZE' and 'PLAYBOOK' sections require the agent to deeply process this untrusted text to extract hooks, formats, and patterns, which could trigger obedience to malicious instructions hidden in the scraped data.\n
  • Sanitization: There is no mention of sanitizing or validating the external content prior to analysis, increasing the risk that embedded prompt injections could influence the agent's output or behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 09:36 AM
Security Audit — agent-trust-hub — social-content