business-location-screening
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to scrape external, untrusted content from the web which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: In
SKILL.md, the agent is directed to usescripts/crawlora.sh /web/scrapeto fetch content from local government pages or user-provided URLs to check for zoning and licensing information. - Boundary markers: The instructions in
SKILL.mdlack explicit delimiters or specific warnings to ignore instructions that might be embedded within the scraped web content. - Capability inventory: The agent has access to various data retrieval tools including Bing search, Google Maps, Apple Maps, and Yelp via the
scripts/crawlora.shhelper, providing a significant surface if the agent is successfully manipulated by external data. - Sanitization: There is no evidence of content sanitization or filtering applied to the results of the
/web/scrapeoperation before it is presented to the agent context.
Audit Metadata