growth-engineer
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines protocols for handling untrusted inputs from web content (via
BrowserOS), which constitutes a potential surface for indirect prompt injection attacks. \n - Ingestion points: Web content and external data processed via
BrowserOS. \n - Boundary markers: No explicit delimiter or 'ignore' instructions are defined in the provided file. \n
- Capability inventory: Includes the ability to execute shell commands through the
ExecutionProxy Interface. \n - Sanitization: Instructions require redacting secrets and PII before context sharing, which is a defensive measure. \n- [COMMAND_EXECUTION]: The skill documentation mandates the use of an
ExecutionProxy Interface(using thertkprefix) to execute terminal actions such asrtk npm test, granting the agent shell access for development tasks.
Audit Metadata