codebase-to-course
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
git clonecommand to retrieve remote repositories and is instructed to open the resulting HTML file in a browser for the user. - [EXTERNAL_DOWNLOADS]: Downloads remote repository data via Git and fetches CSS/font assets from Google Fonts, which is a well-known service.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted codebases. Malicious files within a repository could contain instructions to hijack the agent's logic during the analysis phase. Additionally, the skill explicitly instructs the agent to proceed with generation without seeking user approval for the curriculum, which reduces opportunities for human oversight.
- Ingestion points: Processes all files within a local directory or a cloned repository (Phase 1: Codebase Analysis).
- Boundary markers: No delimiters or safety instructions are defined for the file-reading process.
- Capability inventory: The agent can clone repositories, read filesystem content, and write new files (Phase 3: Build the Course).
- Sanitization: No validation or sanitization of the content extracted from the codebase is performed before processing.
Audit Metadata