Crawl4AI+SearXNG MCP Server
by alexesom
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| HOST | No | Host address for MCP server | 0.0.0.0 |
| PORT | No | Port for MCP server | 8051 |
| NEO4J_URI | No | Neo4j connection URI (required if USE_KNOWLEDGE_GRAPH=true) | bolt://localhost:7687 |
| TRANSPORT | No | Transport protocol for MCP server (sse or stdio) | sse |
| NEO4J_USER | No | Neo4j username (required if USE_KNOWLEDGE_GRAPH=true) | neo4j |
| SEARXNG_TLS | No | Email address for Let's Encrypt certificate (for production HTTPS) | |
| SEARXNG_URL | No | URL of the SearXNG server (pre-configured for Docker Compose) | http://searxng:8080 |
| MODEL_CHOICE | No | LLM model for summaries and contextual embeddings | gemini-2.5-flash-lite |
| SUPABASE_URL | Yes | Supabase project URL for vector database (required) | |
| USE_RERANKING | No | Enable reranking for improved result relevance using cross-encoder models | false |
| GEMINI_API_KEY | Yes | Gemini API key for generating embeddings (required) | |
| NEO4J_PASSWORD | No | Neo4j password (required if USE_KNOWLEDGE_GRAPH=true) | |
| SEARXNG_TIMEOUT | No | Timeout for SearXNG requests in seconds | 30 |
| USE_AGENTIC_RAG | No | Enable specialized code example extraction and storage | false |
| SEARXNG_HOSTNAME | No | Custom domain for production HTTPS deployment | http://localhost |
| USE_HYBRID_SEARCH | No | Enable hybrid search combining vector and keyword search | false |
| SEARXNG_USER_AGENT | No | User agent string for SearXNG requests | MCP-Crawl4AI-RAG-Server/1.0 |
| USE_KNOWLEDGE_GRAPH | No | Enable AI hallucination detection and repository code analysis using Neo4j | false |
| SUPABASE_SERVICE_KEY | Yes | Supabase service key for vector database (required) | |
| SEARXNG_DEFAULT_ENGINES | No | Default search engines for SearXNG | google,bing,duckduckgo |
| USE_CONTEXTUAL_EMBEDDINGS | No | Enable contextual embeddings for enhanced semantic understanding | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |