teach-coding
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands to manage learning projects, including 'npm install' for dependency resolution, 'npm run dev' to launch the Vite development server, and 'bash' to run local test suites ('tests/run.sh'). These operations are necessary for the skill's primary function of interactive programming education.
- [EXTERNAL_DOWNLOADS]: The guide template fetches fonts from Google Fonts and project dependencies from the NPM registry. These are well-known and trusted services used for standard web development.
- [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection as it is designed to ingest and execute user-authored code.
- Ingestion points: The agent reads and processes code files created by the user in the './projects//lab/' directory.
- Boundary markers: No explicit boundary markers or 'ignore' instructions are used when reading user code for verification.
- Capability inventory: The skill has the capability to execute code via the shell, node, and agentic tools like Playwright.
- Sanitization: User-provided code is processed without explicit sanitization or sandboxing instructions beyond the agent's default environment.
Audit Metadata