prompt-skill

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions attempting to override agent behavior or bypass safety filters were detected. The content is strictly focused on UI formatting.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded secrets, or network exfiltration patterns were found. The skill operates locally on user input.
  • [External Downloads] (SAFE): The skill references the 'questionary' package, which is a standard and well-known library for interactive CLI prompts. No unverified remote scripts or piped bash commands are present.
  • [Indirect Prompt Injection] (LOW): The skill ingests untrusted user input via terminal prompts. While this represents a surface for indirect injection, the skill lacks the high-privilege capabilities (like network access or arbitrary shell execution) necessary to weaponize such an attack. Evidence:
  • Ingestion points: questionary.text, questionary.select, and other input functions in SKILL.md.
  • Boundary markers: Standard prompt delimiters are used, though no specific instructions to ignore embedded commands are present.
  • Capability inventory: The skill is restricted to Read/Write operations and UI rendering.
  • Sanitization: Includes basic input validation for titles and dates.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 12:23 PM