site-audit
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests content from external websites, which can act as a vector for indirect instructions.
- Ingestion points: Tools such as
run_site_auditandseo_analyze_pagefetch data from user-specified URLs. - Boundary markers: There are no explicit delimiters or system instructions used to isolate external content from the agent's control logic.
- Capability inventory: The agent has the ability to persist data via
create_taskand publish findings viacreate_shared_report. - Sanitization: The skill documentation does not specify any sanitization or filtering of the website content being analyzed.
Audit Metadata