clean
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests untrusted external data and web content via BrowserOS (SKILL.md).
- Ingestion points: External data and web content accessed via BrowserOS.
- Boundary markers: The instructions lack explicit structural delimiters or explicit instructions to ignore embedded commands for the untrusted content.
- Capability inventory: The skill facilitates code refactoring/modification and terminal execution via the ExecutionProxy Interface (SKILL.md).
- Sanitization: While it mandates redacting secrets/PII before sharing context, there is no specific technical validation mentioned to prevent prompt injection payloads within the external data.
Audit Metadata