thinkies-tutor

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or sensitive data exfiltration attempts were detected in the skill instructions or associated README.
  • [EXTERNAL_DOWNLOADS]: The README provides links to download the skill and installation instructions using a CLI tool ('npx skills add') targeting the author's own GitHub repository ('synapseradio/ai-skills'). These represent standard vendor-provided distribution methods.
  • [INDIRECT_PROMPT_INJECTION]: The skill utilizes the 'WebFetch' tool to gather current information for tutoring. This creates an indirect prompt injection surface where content from the web could potentially influence the agent's behavior.
  • Ingestion points: External data enters the context through the 'WebFetch' tool mentioned in 'SKILL.md'.
  • Boundary markers: No specific delimiters or 'ignore' instructions for external content are defined in the prompt.
  • Capability inventory: The skill's functionality is focused on educational dialogue and planning; it does not utilize sensitive tools or administrative capabilities.
  • Sanitization: There are no explicit sanitization or validation steps for the retrieved web content within the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 03:25 PM
Security Audit — agent-trust-hub — thinkies-tutor