landing-page-copywriting

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core workflow involves ingesting and processing untrusted third-party data such as customer reviews, support tickets, and community forum threads.
  • Ingestion points: External text from reviews and transcripts is brought into the agent context in the 'Mine before you write' step described in SKILL.md and references/copywriting-playbook.md.
  • Boundary markers: The instructions lack specific guidance on using delimiters (e.g., XML tags or triple quotes) or 'ignore embedded instructions' warnings when processing this data.
  • Capability inventory: The skill's primary function is text generation; it does not explicitly request or use capabilities for file writing, system command execution, or network exfiltration.
  • Sanitization: There is no mention of filtering or sanitizing the customer-sourced text before the agent uses it to draft headlines and copy.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 09:55 PM
Security Audit — agent-trust-hub — landing-page-copywriting