qa-automation-engineer

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill incorporates a protocol for processing external data via the BrowserOS tool. Although it mandates defensive measures, this interaction point is a surface for potential indirect prompt injection attacks.
  • Ingestion points: External web content and data processed through BrowserOS.
  • Boundary markers: Explicit instructions to treat external data as 'hostile'.
  • Capability inventory: Command execution via ExecutionProxy and persistence of markdown artifacts via MemoryStore.
  • Sanitization: Requirement to redact secrets and PII before sharing context with subagents.
  • [COMMAND_EXECUTION]: The skill references an ExecutionProxy Interface for executing terminal actions (e.g., rtk npm test). This capability allows the agent to run commands locally, which is an intended function of a QA automation role but requires careful scoping.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 07:36 AM
Security Audit — agent-trust-hub — qa-automation-engineer