landing-page-evaluator

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted user-provided content (existing copy or briefs) which could contain instructions intended to override the agent's logic.
  • Ingestion points: User input and uploaded files are ingested in Step 1 (Intake) to be analyzed and processed.
  • Boundary markers: The skill does not define or use specific delimiters (e.g., XML tags or triple quotes) or 'ignore' instructions to isolate user content from the primary instructions.
  • Capability inventory: The skill possesses the ability to write to the filesystem ('/mnt/user-data/outputs/') and present interactive HTML artifacts to the user.
  • Sanitization: There are no explicit instructions for escaping or sanitizing user-provided text before it is interpolated into the generated HTML wireframe, which could lead to unintended script execution or layout manipulation within the artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 10:48 AM
Security Audit — agent-trust-hub — landing-page-evaluator