What data leaves my machine?
Only embedding vectors (384 floats), the OCIDs you fetch, filter values, and your API key are sent to the API. Your query and profile text stay local.
@qune-tech
基本信息
MCP server for German public procurement data — semantic search and tender matching for Claude, GPT, Cursor, and LM Studio
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"ocds": {
"command": "ocds-mcp",
"args": [
"--api-key",
"sk_live_YOUR_KEY_HERE"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
A local MCP server that connects AI assistants (Claude, GPT, etc.) to the Vergabe Dashboard API for German public procurement. It enables semantic search, tender matching, and company-profile management while keeping query and profile text private on your machine.
Install via npx @qune-tech/vergabe-mcp --api-key sk_live_YOUR_KEY_HERE, download a pre-built binary from GitHub Releases, or build from source. Configure your AI client (Claude Desktop, Claude Code, Cursor, LM Studio) with the command and API key. You can also pass the key via the VERGABE_API_KEY environment variable.
Only embedding vectors (384 floats), the OCIDs you fetch, filter values, and your API key are sent to the API. Your query and profile text stay local.
You need an API key from vergabe-dashboard.qune.de on an Enterprise plan, ~120 MB disk space for the ONNX model (downloaded automatically), and an internet connection to reach the API.
Text embeddings are computed locally with a multilingual ONNX model. Only the resulting vectors are sent to the API, not the original text. Company profiles are stored in a local SQLite database.
On first use the server downloads the model from huggingface.co (~118 MB). No user data is sent in this fetch. For air-gapped installs, place the model files manually in the cache directory.
Yes, the local MCP server is free and open source (MIT license). The API key issuance requires an active Enterprise plan on the Vergabe Dashboard.

Local AI agent for your Mac with a built-in MCP server: 18 tools for your notes and PDFs, a temporal memory graph, an on-device LLM and a neural voice. No account, no telemetry. No data centers required.

Persistent memory for AI agents. Hosted MCP server with capsule-scoped key/value storage, tags, TTL expiration, bulk ops, and RBAC.

Shared memory for AI coding agents. Claude Code, Cursor and Codex read your project's decisions, lessons and tasks before they act.

Open-source Python SDK & MCP interface for Priostack Agent Context Network (ACN). Zero-setup, autonomous memory orchestration for AI agents.

Ask your Takibi Base collections anything; get answers with cited passages and a support score.
评论