ai-engineering-curriculum

Warn

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation includes an installation command npx skills add rohitg00/ai-engineering-from-scratch which retrieves code from an external GitHub repository not associated with the author context.\n
  • Evidence: Located in the 'Overview' section of SKILL.md.\n- [COMMAND_EXECUTION]: The skill suggests the use of the npx skills add command, which downloads and executes logic that modifies the agent's operational environment.\n- [PROMPT_INJECTION]: The skill processes user requests and quiz responses to guide curriculum navigation, representing an indirect prompt injection surface.\n
  • Ingestion points: User queries about starting points and specific AI topics (e.g., 'Where should I start learning AI?').\n
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the documentation.\n
  • Capability inventory: Recommending curriculum modules and conducting knowledge quizzes (non-privileged information processing).\n
  • Sanitization: There is no evidence of input validation or content filtering for user-provided data.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 21, 2026, 07:19 AM
Security Audit — agent-trust-hub — ai-engineering-curriculum