agent-loop-browser
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to visit and process content from external PR preview URLs, which are untrusted ingestion points for data.
- Ingestion points: External websites (PR preview URLs) as described in SKILL.md.
- Boundary markers: The instructions lack markers to delimit or ignore instructions that might be embedded within the web content.
- Capability inventory: The skill utilizes browser tools for navigation, screenshots, and page text extraction.
- Sanitization: There is no instruction for the agent to sanitize or validate content extracted from the pages before reporting it.
- [NO_CODE]: The skill consists entirely of natural language instructions within SKILL.md and does not include any accompanying scripts, executables, or code files.
Audit Metadata