hyperframes

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a website-to-video workflow using the npx hyperframes capture <url> command, which ingests content from arbitrary external websites. This content is then used by the agent to generate design tokens, scripts, and storyboards without explicit sanitization or boundary markers to prevent the ingestion of malicious instructions.
  • [EXTERNAL_DOWNLOADS]: The scripts/setup.sh script performs a global installation of the hyperframes CLI tool from the public npm registry and downloads the chrome-headless-shell binary via Puppeteer. Additionally, the skill includes external JavaScript libraries like GSAP from the cdn.jsdelivr.net CDN.
  • [COMMAND_EXECUTION]: The instructions include commands to terminate system processes using pkill and execute a local setup script via bash. The tool relies extensively on npx for executing various CLI tasks such as capturing, rendering, and media processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 06:05 PM
Security Audit — agent-trust-hub — hyperframes