scraping
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill outlines procedures for scraping external web content, which introduces a potential surface for indirect prompt injection. If the agent processes this untrusted data without sanitization, it could be manipulated by instructions hidden within the target websites.\n
- Ingestion points: External websites via curl, fetch, and browser rendering.\n
- Boundary markers: None specified in the rules to distinguish between data and instructions.\n
- Capability inventory: Includes network fetching, browser rendering, and data storage operations as described in the progressive escalation steps.\n
- Sanitization: No mention of validation, escaping, or filtering for retrieved content.
Audit Metadata