skills/thienqo/ios-skills/quicklook/Gen Agent Trust Hub

quicklook

Warn

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
  • [METADATA_POISONING]: The skill incorrectly labels links to sosumi.ai as "Official Apple Documentation" in the References section of SKILL.md. Official Apple API references are hosted on developer.apple.com. Misrepresenting third-party sources as official is deceptive and can lead developers to trust unverified guidance.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for processing and previewing external local files, creating a vulnerability surface where malicious file content could exploit system-level parsing frameworks.
  • Ingestion points: The fileURL and urls parameters in SKILL.md (DocumentRowView, DocumentPreviewCoordinator) and references/quicklook-patterns.md (QuickLookSheet, ProgressiveThumbnailLoader) accept arbitrary local file paths.
  • Boundary markers: Absent. The skill does not include instructions to use delimiters or security warnings when handling content from untrusted local files.
  • Capability inventory: The skill utilizes QLPreviewController and QLThumbnailGenerator (referenced in SKILL.md and references/quicklook-patterns.md), which involve complex system-level parsing and rendering of document, image, and 3D media content.
  • Sanitization: The QuickLookValidator class in references/quicklook-patterns.md provides basic Uniform Type Identifier (UTType) validation to ensure files match supported formats, but no sanitization of the actual file data is performed.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 9, 2026, 12:02 PM