tw-edu-lesson-plan-108

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a local Python script (scripts/generate_lesson_plan.py) to generate .docx files. It also utilizes system utilities such as soffice (LibreOffice) and pdftoppm (poppler-utils) for document conversion and layout verification.
  • [INDIRECT_PROMPT_INJECTION]: The skill integrates external data from the web using the WebSearch tool to fetch lesson content, word meanings, and teaching materials. This introduces a surface where content from the public internet is processed by the agent.
  • Ingestion points: Web search results (Search A-E) used to populate lesson plan fields.
  • Boundary markers: The prompt uses specific instructions to guide the selection of content, though no technical delimiters are explicitly used when passing search data to the document generator.
  • Capability inventory: The agent has access to Bash (script execution), Write (file creation), and Read (verification).
  • Sanitization: Content is primarily processed by the agent into structured arguments for the generation script, providing a layer of natural language filtering.
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations via the WebSearch tool to retrieve educational resources and curriculum data from the public internet. This is a core functionality required for its intended purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 10:40 AM