k12-core-literacy-planning

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external educational materials, textbook editions, and teacher-supplied documents to generate curriculum content, which creates a surface for indirect instructions to influence the agent's output.
  • Ingestion points: Teacher-supplied materials, textbook editions, and curriculum documents referenced in the 'Ground the lesson before planning' and 'New-stage execution' sections of SKILL.md.
  • Boundary markers: The skill does not define specific delimiters or instructional wrappers (such as 'ignore embedded instructions') for the interpolated data during the generation process.
  • Capability inventory: The skill utilizes platform tools including create_stage, generate_scene, patch_stage, and generate_tts to persist generated content and audio to the classroom environment.
  • Sanitization: The skill attempts to mitigate risk through instructional guidance, requiring the agent to 'pass only verified facts' and use materials for scope rather than direct reproduction.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 10:40 AM
Security Audit — agent-trust-hub — k12-core-literacy-planning