codebase-to-course
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external codebases (via local paths or remote Git repositories), which presents a surface for indirect prompt injection if the code contains malicious instructions in comments, documentation, or string literals.
- Ingestion points: Remote Git repositories or local directories provided by the user (SKILL.md).
- Boundary markers: Absent. The skill instructions do not direct the agent to treat the analyzed content as untrusted or to ignore embedded instructions.
- Capability inventory: The skill can read files, execute
git clone, and write an HTML file containing JavaScript (SKILL.md). - Sanitization: Absent. The instructions do not specify sanitization or escaping of code snippets extracted from the codebase before inclusion in the final output.
- [COMMAND_EXECUTION]: The skill executes shell commands, specifically
git clone, using user-provided URLs to download repositories for analysis (SKILL.md). - [DYNAMIC_EXECUTION]: The skill generates a self-contained HTML file containing JavaScript code for interactive UI elements such as animations and tooltips. This script generation is based on patterns and templates provided in the reference files (SKILL.md, references/interactive-elements.md).
- [EXTERNAL_DOWNLOADS]: The skill performs network operations to fetch content from external Git repositories specified by the user (SKILL.md). It also references Google Fonts, a well-known service, for typography in the generated output.
Audit Metadata