structured-delivery

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • SAFE (SAFE): The skill consists entirely of markdown instructions for the AI to format its output. It does not include any scripts, binaries, or configuration files that execute logic.
  • Indirect Prompt Injection (LOW): The skill defines a surface for processing user-provided information into structured formats. Ingestion points: User input triggers activation gates in SKILL.md. Boundary markers: None. Capability inventory: No file access, network operations, or subprocess execution capabilities exist in this skill. Sanitization: None.
  • Prompt Injection (SAFE): No instructions were found that attempt to bypass safety filters or override core agent constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 11:21 AM