SearXNG
MCP ServerFree** - A Model Context Protocol Server for [SearXNG](https://docs.searxng.org)
Capabilities5 decomposed
mcp-native metasearch query execution
Medium confidenceExecutes search queries against a SearXNG instance through the Model Context Protocol, translating MCP tool-call schemas into HTTP requests to SearXNG's REST API and marshaling results back as structured JSON. The implementation wraps SearXNG's `/search` endpoint with MCP's standardized tool-calling interface, enabling LLM agents to invoke searches without direct HTTP knowledge.
Bridges SearXNG (a privacy-respecting metasearch engine) with MCP protocol, enabling LLM agents to use decentralized search without relying on commercial search APIs. Unlike direct HTTP integration, MCP standardization allows any MCP-compatible client (Claude, custom agents) to use the same interface.
Provides privacy-first search integration for MCP agents without vendor lock-in to OpenAI/Google APIs, though with lower result quality than commercial search engines due to SearXNG's aggregation model
searxng parameter mapping and schema translation
Medium confidenceTranslates MCP tool-call parameters into SearXNG-compatible query parameters, handling schema validation, parameter normalization, and optional argument handling. The server maintains a mapping layer between MCP's standardized tool schema and SearXNG's query API, including support for filters like language, category, and time range if the SearXNG instance exposes them.
Implements a declarative parameter mapping layer that abstracts SearXNG's query API behind MCP's tool schema, allowing clients to remain agnostic to SearXNG's specific parameter names and formats while maintaining type safety.
More maintainable than hardcoding SearXNG parameters directly in agent prompts, and more flexible than generic HTTP wrappers because it validates parameters before execution
multi-engine result aggregation and normalization
Medium confidenceReceives heterogeneous search results from SearXNG's aggregated engines (Google, Bing, DuckDuckGo, etc.) and normalizes them into a consistent JSON schema with fields like title, URL, snippet, and source engine. The normalization layer handles varying result formats from different search engines and presents a unified interface to MCP clients.
Normalizes results from SearXNG's multi-engine aggregation into a single schema, preserving source attribution so clients can trace which engine provided each result — useful for privacy audits and result quality analysis.
More transparent than opaque search APIs because it exposes which engine returned each result, enabling agents to make informed decisions about result trustworthiness
self-hosted searxng instance configuration and discovery
Medium confidenceAccepts configuration parameters (SearXNG instance URL, optional authentication credentials) to connect to a specific SearXNG deployment, with optional auto-discovery of instance capabilities via SearXNG's `/config` endpoint. The server can detect available search engines, supported languages, and categories from the target instance, adapting its tool schema dynamically.
Supports dynamic discovery of SearXNG instance capabilities via the `/config` endpoint, allowing the MCP server to adapt its tool schema to match the actual engines and languages available on the target instance rather than assuming a fixed configuration.
More flexible than hardcoded SearXNG configurations because it auto-detects capabilities, enabling the same MCP server to work with different SearXNG deployments without code changes
mcp protocol compliance and tool schema exposition
Medium confidenceImplements the Model Context Protocol specification for tool servers, exposing search capabilities as standardized MCP tools with JSON Schema definitions. The server registers tool definitions with required and optional parameters, handles MCP tool-call requests, and returns results in MCP's expected format, enabling seamless integration with any MCP-compatible client.
Fully implements MCP protocol specification, exposing SearXNG as a standardized tool that any MCP-compatible client can discover and invoke without custom integration code. This enables SearXNG to be composed with other MCP tools in a unified agent architecture.
More interoperable than custom HTTP wrappers because it uses a standard protocol that multiple clients already support, reducing integration friction
Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.
Related Artifactssharing capabilities
Artifacts that share capabilities with SearXNG, ranked by overlap. Discovered automatically through the match graph.
mcp-searxng
MCP server for SearXNG integration
serpapi-mcp
MCP server: serpapi-mcp
Meilisearch
** - Interact & query with Meilisearch (Full-text & semantic search API)
miyami-websearch-mcp
MCP server: miyami-websearch-mcp
SearXNG
Privacy-respecting metasearch — 70+ engines, no tracking, self-hosted, JSON API for AI agents.
SearXNG Search Server
Provide programmatic access to privacy-respecting meta-search functionality via a standardized protocol. Perform advanced search queries with flexible filtering and output formats. Easily deploy and integrate with existing SearXNG instances using multiple transport modes including HTTP and stdio.
Best For
- ✓AI agent developers building privacy-first applications
- ✓Teams running self-hosted SearXNG instances who want LLM integration
- ✓Builders implementing MCP servers for closed-loop AI systems
- ✓MCP server developers building search integrations
- ✓Teams standardizing on MCP for tool-calling across heterogeneous backends
- ✓Agents that need consistent result parsing across multiple search sources
- ✓Applications requiring transparency about result provenance
- ✓Teams running self-hosted SearXNG with varying configurations
Known Limitations
- ⚠Depends on SearXNG instance availability — no fallback to public search engines if instance is down
- ⚠Search result quality and coverage varies by SearXNG configuration and enabled search engines
- ⚠No built-in result ranking or deduplication — returns raw aggregated results from SearXNG
- ⚠Latency depends on SearXNG instance performance and network distance
- ⚠Parameter support is limited to what SearXNG exposes — advanced filters may not be available
- ⚠No parameter validation against SearXNG instance capabilities at runtime — assumes standard SearXNG configuration
Requirements
Input / Output
UnfragileRank
UnfragileRank is computed from adoption signals, documentation quality, ecosystem connectivity, match graph feedback, and freshness. No artifact can pay for a higher rank.
About
** - A Model Context Protocol Server for [SearXNG](https://docs.searxng.org)
Categories
Alternatives to SearXNG
Search the Supabase docs for up-to-date guidance and troubleshoot errors quickly. Manage organizations, projects, databases, and Edge Functions, including migrations, SQL, logs, advisors, keys, and type generation, in one flow. Create and manage development branches to iterate safely, confirm costs
Compare →Are you the builder of SearXNG?
Claim this artifact to get a verified badge, access match analytics, see which intents users search for, and manage your listing.
Get the weekly brief
New tools, rising stars, and what's actually worth your time. No spam.
Data Sources
Looking for something else?
Search →