tweet-insight

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified through the automated crawling of external links found in tweets.
  • Ingestion points: The skill uses browser_navigate and WebFetch to read content from user-provided tweets and any external links (articles, papers, blogs) discovered within them.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are used when interpolating external content into the context.
  • Capability inventory: The agent has access to browser tools (browser_navigate, browser_snapshot) and web fetching capabilities.
  • Sanitization: There is no evidence of sanitization or filtering of the retrieved content before it is digested and processed by the LLM.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves content from the live web.
  • Fetches data from external sites using r.jina.ai, a well-known web-to-markdown service.
  • Uses Playwright MCP tools to navigate and snapshot social media platforms.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 02:10 PM