competitive-ads-extractor
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to ingest and analyze untrusted text from third-party advertisements on platforms like Facebook and LinkedIn.
- Ingestion points: Untrusted data enters the context from external ad libraries as described in the scraping workflow in SKILL.md.
- Boundary markers: The skill does not provide clear delimiters or instructions to ignore embedded commands within the analyzed ad copy.
- Capability inventory: The skill requires the agent to use web browsing/scraping tools and file system write access to save analysis to directories like '~/competitor-ads/'.
- Sanitization: There is no mention of sanitization or filtering of the retrieved ad content before processing.
- [NO_CODE]: The provided skill contains only natural language instructions and markdown documentation. It does not include executable code, scripts, or external dependencies, relying entirely on the host agent's native tools.
Audit Metadata