learn
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches lesson content, quizzes, and course manifests from the author's repository on GitHub (rohitg00/ai-engineering-from-scratch).
- [INDIRECT_PROMPT_INJECTION]: The skill processes curriculum data from external sources, which constitutes a potential injection surface.
- Ingestion points: Reads local
LEARNING.mdprogress files and fetches Markdown (en.md) and JSON (quiz.json) lesson data from a remote repository. - Boundary markers: No explicit boundary markers or 'ignore' instructions are used when interpolating lesson content into the agent's context.
- Capability inventory: The skill facilitates shell command execution for course navigation and is designed to execute code snippets from the lessons.
- Sanitization: Lesson and quiz content is processed without specific sanitization or validation logic.
- [COMMAND_EXECUTION]: The skill directs the agent to use platform-specific commands (such as slash commands in Claude Code) to manage the learning flow and check understanding.
- [DYNAMIC_EXECUTION]: To demonstrate concepts, the skill is instructed to run code chunks from the fetched lessons and present the real output to the learner.
Audit Metadata