technical-diagram-analyzer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted visual content and diagrams while maintaining high-privilege tool access without defined safety boundaries.
  • Ingestion points: Processes user-provided technical diagrams, charts, and visual documentation tools (SKILL.md).
  • Boundary markers: Absent. There are no instructions for the agent to distinguish between its own system instructions and instructions potentially embedded in the diagrams or code being analyzed.
  • Capability inventory: Access to Bash, Write, Edit, Read, and Grep. This combination allows for significant system modification and command execution (SKILL.md).
  • Sanitization: Absent. The skill does not define validation or escaping protocols for data extracted from visual inputs before processing them through high-privilege tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:32 PM