assetseeker

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the discovery and download of assets from established and well-known services such as Pexels, Unsplash, Pixabay, Iconify, and Google Fonts, which is the primary intended purpose of the skill.
  • [DATA_EXFILTRATION]: The skill correctly uses environment variables (e.g., PEXELS_API_KEY, UNSPLASH_ACCESS_KEY) for API authentication, following security best practices by avoiding hardcoded credentials and relying on local configuration.
  • [PROMPT_INJECTION]: While the skill ingests external data (asset metadata and descriptions) from third-party sources, the risk of indirect prompt injection is mitigated by the curated nature of the supported sources and the use of standard parsing techniques.
  • Ingestion points: scripts/seek_assets.py via urllib.request.urlopen and Playwright-based web scrapers.
  • Boundary markers: The skill presents data in a structured format (JSON) or plain text without specific delimiters for the AI agent.
  • Capability inventory: The skill has file-writing capabilities via the download command and network access for API queries.
  • Sanitization: Input keywords are sanitized using urllib.parse.quote to prevent URI injection.
  • [SAFE]: The Python implementation uses standard libraries for most functionality, with Playwright as an optional dependency for scraping tasks on legitimate creative resource sites. The behavior is consistent with the skill's stated purpose and author information.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 01:57 AM
Security Audit — agent-trust-hub — assetseeker