Schema | Crawl4AI+SearXNG MCP Server

Crawl4AI+SearXNG MCP Server

Describes the environment variables required to run the server.

Name	Required	Description	Default
`USE_QDRANT`	No	Use Qdrant instead of Supabase (optional)	false
`SEARXNG_URL`	No	Internal SearXNG URL	http://searxng:8080
`CRAWL4AI_URL`	No	Internal Crawl4AI URL	http://crawl4ai:8000
`LLM_PROVIDER`	No	The LLM provider to use (e.g., openai, anthropic, groq)	openai
`USE_SUPABASE`	No	Enable Supabase for vector storage	true
`USE_RERANKING`	No	Cross-encoder Reranking (improves result relevance)	false
`OPENAI_API_KEY`	No	Required for contextual embeddings
`RERANKING_MODEL`	No	The model to use for reranking	cross-encoder/ms-marco-MiniLM-L-12-v2
`USE_AGENTIC_RAG`	No	Enable advanced RAG features	true
`NEO4J_BATCH_SIZE`	No	Batch size for large repository processing	50
`REPO_MAX_SIZE_MB`	No	Maximum repository size	500
`USE_HYBRID_SEARCH`	No	Hybrid Search (combines vector + keyword search). Requires PostgreSQL full-text search	false
`NEO4J_BATCH_TIMEOUT`	No	Timeout for batch operations	120
`REPO_MAX_FILE_COUNT`	No	Maximum number of files	10000
`USE_KNOWLEDGE_GRAPH`	No	Enable Neo4j for code analysis	true
`CRAWL4AI_MAX_CONCURRENT`	No	Maximum concurrent crawling sessions	20
`SUPABASE_MAX_CONNECTIONS`	No	Maximum connections for Supabase	20
`USE_CONTEXTUAL_EMBEDDINGS`	No	Contextual Embeddings (improves search accuracy by 20-30%). Requires OpenAI API or compatible LLM	false

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.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed