legal-test-builder-patrick-munro
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The generated HTML artifacts include a reference to Google Fonts (fonts.googleapis.com). This is a well-known service for web typography and is documented as the sole external dependency for the skill.
- [COMMAND_EXECUTION]: The skill instructions and provided reference files contain JavaScript for client-side functionality within the generated HTML. This includes logic for a countdown timer using
setInterval, hover-based tooltips for document annotations, and UI toggles for reveal blocks. These operations are local to the browser environment and do not involve system-level command execution. - [DATA_EXFILTRATION]: While the generated assessments include textareas for user input, the provided scripts do not contain any network operations or
fetch()calls to transmit data to external servers. The skill focuses on creating 'offline-capable' and 'self-contained' files. - [PROMPT_INJECTION]: The skill follows a logical, step-by-step process for gathering inputs and designing legal tasks. There are no patterns suggesting attempts to bypass AI safety guardrails or override system instructions.
- [SAFE]: The code and instructions are transparent, well-documented, and consistent with the stated purpose of building educational and assessment tools. No obfuscation, persistence mechanisms, or privilege escalation attempts were found.
Audit Metadata