beautiful-html

Warn

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches HTML templates from a third-party GitHub repository (github.com/zarazhangrui/beautiful-html-templates) that is not an established or well-known service.
  • [COMMAND_EXECUTION]: Executes npx playwright install chromium --with-deps, which downloads external binaries and attempts to install system-level dependencies, potentially requiring elevated privileges.
  • [COMMAND_EXECUTION]: Uses node -e to execute a dynamically generated script that interacts with the file system to render and save PDF files.
  • [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by ingesting untrusted user content and interpolating it into HTML templates without sanitization. 1. Ingestion points: User-provided text or file content read during Phase 1. 2. Boundary markers: None identified; user content is processed without clear delimiters. 3. Capability inventory: File reading, network fetching (WebFetch), shell command execution (npx, node), and local file writing. 4. Sanitization: None identified; the skill performs direct placeholder replacement in Phase 4.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 11, 2026, 10:31 AM
Security Audit — agent-trust-hub — beautiful-html