competitive-ads-extractor
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted data from external ad libraries, creating an attack surface for indirect prompt injection.
- Ingestion points: Scraped ad copy and visuals from external platforms like Facebook and LinkedIn (SKILL.md).
- Boundary markers: Absent. There are no instructions for the agent to use delimiters or ignore instructions found within the scraped content.
- Capability inventory: The skill utilizes file system write capabilities to save analysis reports and screenshots (e.g., in ~/competitor-ads/).
- Sanitization: Absent. The skill does not specify any validation or filtering for the external data before it is processed by the model.
- [NO_CODE]: This skill consists solely of markdown instructions and examples. It does not contain any scripts, binaries, or automated configuration files.
Audit Metadata