thumbnail-of-pptx

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by navigating to and rendering untrusted content from external URLs.
  • Ingestion points: The url argument in thumbnail_of_pptx.py used in page.goto(url).
  • Boundary markers: Absent. No specific delimiters or safety instructions are provided to the agent regarding the content of the captured slide.
  • Capability inventory: The skill uses Playwright to launch a browser, navigate to network resources, and write screenshots to the local file system.
  • Sanitization: The input URL is not validated or sanitized, allowing it to point to any domain or local protocol.- [COMMAND_EXECUTION]: The script uses the Playwright library to launch and control a headless Chromium browser instance.- [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install Chromium browser binaries using the playwright install command.- [DATA_EXFILTRATION]: Potential for local data exposure if the provided URL uses the file:// protocol, allowing the agent to screenshot local files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 09:16 PM
Security Audit — agent-trust-hub — thumbnail-of-pptx