Ad Creative Strategist
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to ingest and analyze external data from advertising platforms.
- Ingestion points: Processes existing ad copy and performance data (e.g., via list_ads).
- Boundary markers: None identified in the prompt instructions to isolate external data.
- Capability inventory: Uses Bash, WebFetch, Write, and Edit tools.
- Sanitization: No content validation or sanitization of ingested ad copy is specified.
Audit Metadata