zohaamalik/ocds-procura-semantic-nexus

OCDS-MCP is an open-source framework to connect AI assistants to German public procurement (OCDS) data. It ingests, normalizes and vectorizes tender records, exposing them via a Model Context Protocol (MCP) API so LLMs can perform semantic search, intelligent tender matching and generate private company profiles without manual data wrangling. Who uses it: AI/ML developers, procurement analysts, public-sector transparency teams, competitive-intelligence groups and researchers. Key features: data ingestion and normalization, transformer embeddings (OpenAI text-embedding-3-small; optional Claude), semantic tender search, similarity matching, private company profiler with bidding history, MCP-compliant JSON output, React UI, multilingual DE/EN support, batch processing and modular/scalable architecture. Runs with Python 3.10+ or Node.js; requires API keys for embedding/reasoning services.

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.