eli5-figure
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads fonts from jsDelivr and software packages from the npm registry and PyPI (via
uvx). These are well-known, trusted services. The setup script (scripts/setup.sh) implements security best practices by verifying the SHA-256 hashes of downloaded fonts before use. - [COMMAND_EXECUTION]: The skill executes local Node.js and shell scripts to perform graph layout and rendering tasks. This is the primary function of the skill and does not involve arbitrary or malicious command execution.
- [DYNAMIC_EXECUTION]: The skill uses Playwright to launch a headless Chromium browser for rendering SVG content and MathJax. It also uses
elkjsfor runtime graph layout calculations. These operations are restricted to the local environment and are necessary for the skill's purpose. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied graph definitions in JSON format. While this presents an ingestion surface for untrusted data, the skill utilizes standard parsers (e.g.,
JSON.parse,DOMParser) and limits its operations to diagram generation, which significantly restricts the potential for exploitation.
Audit Metadata