teach-wx
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a structured pedagogical assistant, maintaining a local workspace for educational content and progress tracking. All file operations are scoped to the project directory and intended for educational record-keeping.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes external information and user input to generate lessons. This is a functional requirement of its intended purpose and is evaluated as a low-risk factor. • Ingestion points: External sources identified during research and user-defined topics (SKILL.md). • Boundary markers: Not explicitly defined in the provided HTML or Markdown templates to separate external data from instructions. • Capability inventory: Local file system writes to specific subdirectories for lesson, record, and glossary storage (SKILL.md). • Sanitization: No explicit content filtering or escaping for external data is specified in the teaching logic.
Audit Metadata