kangyongcai

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process untrusted external data, such as customer emails or student messages, to generate advice and draft replies. This creates a potential surface for indirect prompt injection attacks.\n
  • Ingestion points: The skill ingests user-provided text through its primary communication modes (A, B, and C) as defined in SKILL.md.\n
  • Boundary markers: There are no explicit boundary markers or instructions to treat the input strictly as data, which could allow the agent to be influenced by instructions embedded within the processed messages.\n
  • Capability inventory: The skill does not request specific tools or perform high-risk operations like shell command execution or network requests, which significantly mitigates the potential impact of any injection.\n
  • Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the external content before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 07:32 AM
Security Audit — agent-trust-hub — kangyongcai