searching-web

Warn

Audited by Snyk on Mar 29, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.90). This skill explicitly performs live web searches and URL crawling of open/public sources (INSTRUCTIONS.md shows web_search_advanced_exa with categories including "tweet"/"personal site" and the crawling_exa({url: ...}) example) and the fallback scripts/web-search.sh runs a gemini CLI search that fetches and summarizes external webpages, so the agent ingests untrusted third‑party content that can influence its subsequent decisions.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 29, 2026, 11:14 PM
Issues
1