refine

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting untrusted data from an external source.
  • Ingestion points: Ticket data is retrieved from a backlog tool via the 'bd show' command.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the retrieved data.
  • Capability inventory: The skill can create new records via 'bd create' and delegate logic to sub-agents.
  • Sanitization: No explicit validation or filtering of input data is defined.
  • Mitigation: The skill enforces an 'INTERACTIVE MODE ONLY' constraint with mandatory stops after each phase, ensuring human oversight before any automated actions are taken.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 05:54 AM
Security Audit — agent-trust-hub — refine