w-debug-browser
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent interacts with external URLs and processes runtime evidence such as DOM content, console logs, and network traces. This external data could contain malicious instructions designed to subvert the agent's logic.
- Ingestion points: External URLs and browser runtime evidence (DOM, network, console) specified in SKILL.md.
- Boundary markers: The instructions do not provide explicit delimiters or "ignore embedded instructions" warnings for the agent when processing external browser data.
- Capability inventory: The workflow implies capabilities for network interaction via browser tools and file system modifications when applying fixes (SKILL.md).
- Sanitization: There are no instructions for sanitizing or validating the ingested runtime evidence before the agent acts upon it.
Audit Metadata