Content Personalization

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTIONNO_CODE
Full Analysis
  • Prompt Injection (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) due to its core function of processing untrusted external data.
  • Ingestion points: The skill ingests 'textbook content' and 'user profile' data as primary inputs for adaptation.
  • Boundary markers: Absent. The instructions do not define delimiters or specify that the agent should ignore instructions embedded within the textbook content.
  • Capability inventory: None. The skill consists entirely of natural language instructions and lacks scripts, network access, or file-writing capabilities.
  • Sanitization: None. There is no logic provided to filter or escape malicious instructions that might be hidden in the content being personalized.
  • Data Exposure (INFO): The workflow requires retrieving information from the user profile. While necessary for the stated purpose, this creates a data exposure surface if the agent is manipulated into revealing more profile information than intended during the adaptation process.
  • No Code (INFO): The skill contains no executable scripts, binaries, or configuration files, which significantly limits the potential for technical exploitation like RCE or persistence.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 05:11 AM