stitched-full-page-capture

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The helper script scripts/stitch_full_page_capture.mjs executes the system tools ffmpeg and sips using subprocess calls (spawn and execFile). These tools are used to process image segments and verify dimensions based on parameters provided in the manifest file.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection through its ingestion of the manifest.json file.
  • Ingestion points: The script processes pageUrl, fullPageImage, and sectionImages arrays from the manifest file provided at runtime.
  • Boundary markers: There are no boundary markers or instructions to the agent to ignore potentially malicious content within the manifest data.
  • Capability inventory: The skill has the capability to visit arbitrary URLs via Playwright and write/read files on the local filesystem using ffmpeg and standard Node.js filesystem modules.
  • Sanitization: While basic sanitization is applied to directory names generated from titles, the file paths extracted from the manifest (e.g., fullPageImage and section.file) are joined into target paths without explicit validation against path traversal (e.g., ../ sequences).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 06:56 PM
Security Audit — agent-trust-hub — stitched-full-page-capture