bookmarks-management

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill facilitates the extraction of extensive personal data from X/Twitter bookmarks, including private content, media, and metadata, which is then exported to local files. While the export is local, the large-scale harvesting of user data represents a potential exposure risk if the resulting files are handled insecurely by the agent.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting untrusted content from social media posts.
  • Ingestion points: scripts/scrapeBookmarks.js and src/bookmarkOrganizer.js ingest tweet text via the [data-testid="tweetText"] selector.
  • Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the documentation.
  • Capability inventory: The skill includes file system writes (CSV/JSON exports) and browser automation via Puppeteer (src/bookmarkManager.js).
  • Sanitization: There is no mention of sanitizing or validating the scraped tweet content before processing or exporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 09:53 AM