start-learning
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a curriculum data file from the author's repository at
raw.githubusercontent.com/rohitg00/ai-engineering-from-scratch/main/ROADMAP.md. This is a vendor-owned resource used to populate estimated study hours. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input from interview questions and external curriculum metadata, which is then persisted in a file for use by other tutor skills.
- Ingestion points: User-provided interview answers (goals and reasons for learning) and content from the remote
ROADMAP.mdfile. - Boundary markers: The generated
LEARNING.mdfile includes markdown comments identifying it as a managed file. - Capability inventory: The skill writes a local markdown file and triggers subsequent curriculum-related skills.
- Sanitization: There is no explicit sanitization step mentioned for interpolating user responses into the tracking file.
- [DYNAMIC_EXECUTION]: The skill dynamically assembles the
LEARNING.mdfile at runtime by combining a predefined template with user-provided intent and remote curriculum data. This file serves as the stateful context for all future learning sessions.
Audit Metadata