searxng-search

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow calls a (potentially public) SearXNG instance via curl to fetch JSON search results (${SEARXNG_URL}/search?...&format=json), and those results include outsider-authored snippets/titles/metadata from external web sources that are ingested into the agent context for further extraction.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill instructs self-hosting steps that modify system files (e.g., "Edit /etc/searxng/settings.yml"), install system-level packages, and run Docker with host volumes — all actions that change the machine state and typically require elevated privileges.