Welcome! This is a Model Context Protocol (MCP) server that exposes the internal Semantic Croissant dataset catalog.
search_croissant_datasets(q, limit, page): Search for datasets using natural language keywords (e.g. "climate change vietnam"). It leverages QLever's internal ranking system to find the most relevant datasets.get_croissant_dataset(id): Retrieve the full, detailed JSON-LD Metadata catalog for a specific dataset ID./expert/.index at 0x75ae52765f80> : Directly query specialized Elasticsearch indices (croissant, dataverse, ollama, huggingface, openml, hips) using standard HTTP requests (e.g. /expert/croissant/_search).This server provides an SSE (Server-Sent Events) transport for MCP.
SSE Endpoint: https://mcp.dev.codata.org/mcp/sse
Powered by QLever and the standard python MCP SDK.