SearXNG Local

**Quick judgement** A solid, privacy-focused web search skill that leverages a self-hosted SearXNG instance. Technically straightforward (HTTP+JSON) and flexible, but it depends heavily on the user correctly deploying and maintaining SearXNG. Best suited for technical users or organizations already comfortable with Docker and self-hosted services. **Key strengths** - Uses a **self-hosted SearXNG** instance, keeping search logs and traffic under the user’s control. - Aggregates results from **many search engines** (Google, DuckDuckGo, Brave, Startpage, etc.) with support for categories, language/region, and pagination. - **No external API keys or SaaS dependencies**, reducing vendor lock-in and rate-limit issues. **Risks / limitations** - **Operational burden**: the user must deploy, configure, and keep SearXNG running (Docker, settings.yml, networking), which is non-trivial for non-technical users. - **Reliability depends on upstream engines**: some engines may be rate-limited or blocked, leading to intermittent or inconsistent results. - **Security & privacy hinge on configuration**: exposing SearXNG on `0.0.0.0` without proper firewall/TLS could leak search traffic over the network; misconfigured engines or proxies may still leak data to third parties. - **Performance variability**: SearXNG may be slower than single-engine APIs, especially when aggregating many sources or under network constraints. **Recommended scenarios** - Privacy-conscious individuals or teams who are comfortable running Docker and want full control over their search stack. - Organizations needing a **centralized, policy-controlled search service** for agents, possibly restricted to certain engines or internal resources. - Agent workflows that perform **heavy or frequent web research** and would otherwise hit commercial API limits or cost ceilings. - Developers building automation or research bots that must avoid vendor lock-in and prefer open-source, self-hosted infrastructure.