novel-art

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The script scripts/novel-art.mjs and scripts/selftest.mjs use only Node.js standard libraries (node:fs, node:path, node:url, node:assert). No external npm packages or remote scripts are downloaded or executed.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool for local script execution. These scripts perform deterministic data validation and rendering tasks without incorporating unvalidated user input into shell commands.
  • [DATA_EXFILTRATION]: No network operations to external or untrusted domains were detected. The skill generates local reports and uses a secure blob download mechanism in the browser for exporting data.
  • [DYNAMIC_EXECUTION]: The HTML report generator in scripts/novel-art.mjs correctly sanitizes model-generated data using an escaping function (esc) and prevents script block breakouts by escaping the < character in embedded JSON data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 11:39 AM
Security Audit — agent-trust-hub — novel-art