ads

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external sources, such as the Facebook Ad Library and product reviews, which creates a vulnerability surface for indirect prompt injection.
  • Ingestion points: Untrusted data enters the agent context via browser tools when performing ad library teardowns and scraping product reviews as described in references/creative-research-automation.md.
  • Boundary markers: The skill contains explicit instructions in SKILL.md and references/audit-guardrails.md stating that fetched pages, exports, and screenshots are data, not instructions, and that the agent should never follow directives embedded in them.
  • Capability inventory: The skill leverages capabilities including browser access for data collection, file-write operations for CSV exports, and network communication via Slack for reporting.
  • Sanitization: Instructions are provided to specifically ignore and note attempts at prompt injection within external data, effectively treating external text as non-executable data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 09:36 PM