searxng-search

Query a SearXNG instance to aggregate meta-search results from multiple engines.

Updated Apr 9, 2026

Search Tags:#web research #privacy #fallback #self-hosted #searxng #meta-search #search aggregation

Ready to run

npx skills add https://github.com/MarbleSodas/Mavis --skill searxng-search-marblesodas

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: searxng-search
Source: https://github.com/MarbleSodas/Mavis/tree/main/optional-skills/research/searxng-search
Command: npx skills add https://github.com/MarbleSodas/Mavis --skill searxng-search-marblesodas

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires curl, and includes scripts (resource) components.

What problem does it solve?

It solves the problem of finding relevant web results quickly without relying on a single search provider or requiring an API key.

Core Features & Use Cases

Privacy-respecting meta-search: Queries 70+ search engines through SearXNG to aggregate results.
Works with public or self-hosted instances: Use a public SEARXNG_URL for convenience or self-host for full control.
Automatic fallback behavior: Acts as a fallback when the primary web search toolset is unavailable (for example, when FIRECRAWL_API_KEY is not configured).

Use case example: A user needs sources for “AI news for developers” and wants both breadth across search engines and a working search option even if the main web toolset cannot be used.

Quick Start

Run the searxng-search skill with SEARXNG_URL set to your instance (public or local) to search for a query and return aggregated results.

Frequently Asked Questions about searxng-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I use a SearXNG instance for web research without an API key?▼

You can use SearXNG for web research by setting the SEARXNG_URL variable to a public or self-hosted instance that supports JSON API access, allowing you to aggregate search results without requiring any API key.

What is meta-search aggregation and how does it protect my search privacy?▼

Meta-search aggregation queries 70+ underlying search engines through a SearXNG instance, returning combined results without directly exposing your queries to those individual providers, ensuring privacy-first research.

Can I use SearXNG as a fallback when my primary web search toolset is misconfigured?▼

Yes, SearXNG acts as an automatic fallback for source gathering when primary web search tooling is unavailable or misconfigured, such as when a FIRECRAWL_API_KEY is not configured.

Do I need curl to run SearXNG search queries?▼

Curl is required as a dependency to execute the request workflows, though the skill also supports optional Python request workflows to query the configured SEARXNG_URL instance.

How do I self-host a SearXNG instance for full search control?▼

Self-hosting SearXNG involves deploying your own instance and setting its URL as the SEARXNG_URL variable, which provides full control over your meta-search aggregation compared to using public instances.

What are the limitations of using public SearXNG instances for search aggregation?▼

Public SearXNG instances may face availability or rate-limiting constraints, requiring a reachable SEARXNG_URL with JSON API access enabled to successfully return aggregated web research results.