deploy
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external inputs from web content and BrowserOS, which introduces a potential surface for indirect prompt injection attacks.
- Ingestion points: External web content and data via BrowserOS mentioned in SKILL.md.
- Boundary markers: Explicit markers are absent, although instructions mandate treating external inputs as hostile.
- Capability inventory: The skill utilizes an ExecutionProxy (rtk prefix) for terminal actions and a MemoryStore interface for writing persistent markdown artifacts to Obsidian documentation.
- Sanitization: Mandatory redaction of secrets and PII is required before external data is shared within the agent context.
Audit Metadata