qa

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by fetching and processing untrusted content from external websites defined in the data.
  • Ingestion points: The agent fetches origin HTML pages via HTTP in Phase 2.
  • Boundary markers: No specific boundary markers or instruction delimiters are mentioned for the external content.
  • Capability inventory: The skill uses Bash, Write, Edit, and Grep tools.
  • Sanitization: The skill requires user confirmation (AskUserQuestion) before applying any fixes and uses a structured parseContent step.
  • [EXTERNAL_DOWNLOADS]: The skill performs automated HTTP requests to external source URLs to retrieve content for quality comparison.
  • [COMMAND_EXECUTION]: The skill invokes the Bash tool and executes local JavaScript/TypeScript logic (e.g., runQa) to automate the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 09:44 PM
Security Audit — agent-trust-hub — qa