geo-audit
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for website analysis and reporting. It uses standard tool calls to fetch public web content and does not attempt to access sensitive local files, environment variables, or hardcoded credentials.
- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection via external data ingestion. 1. Ingestion points: Website content, key pages, and robots.txt (Steps 1, 2, and 4). 2. Boundary markers: None. 3. Capability inventory: Web search and fetching tools only. 4. Sanitization: None. This is a common characteristic of web-browsing skills and is categorized as low risk given the skill's analytical nature.
Audit Metadata