Bidding Big Data

You can run and use a dedicated MCP server that provides comprehensive bidding and tender data services. It supports fuzzy enterprise searches, bid and tender statistics, procurement analytics, and detailed bidding information, enabling you to analyze opportunities and monitor market activity.

How to use

You connect to the MCP server from your MCP client to perform data queries and analytics. Use the available endpoints to search for companies, view bid wins, explore tender statistics, inspect procurement data, query bidding and planned projects, and search for specific招投标 information. You can run the server locally for development or connect to a remote MCP URL to access the hosted service. When using the client, select the appropriate server connection method (HTTP/remote or local STDIO) and provide any required authentication tokens or environment variables as configured.

How to install

Prerequisites: Python 3.10 or newer, and the required Python packages listed in the project dependencies.

Step 1: Clone the project and enter the directory.

git clone https://github.com/handaas/bidding-mcp-server
cd bidding-mcp-server

Step 2: Create a virtual environment and install dependencies.

python -m venv mcp_env && source mcp_env/bin/activate
pip install -r requirements.txt

Step 3: Configure environment variables for the MCP server.

cp .env.example .env

Then edit the .env file to include your integration details.

Example environment variables to configure.

INTEGRATOR_ID=your_integrator_id SECRET_ID=your_secret_id SECRET_KEY=your_secret_key

Step 4: Start the server in a streamable HTTP mode (recommended for development). You will be served at http://localhost:8000.

python server/mcp_server.py streamable-http

## How to connect via MCP client configurations

You can connect using either a remote HTTP URL or a local STDIO command configuration. Use the following examples as starting points in your MCP client setup.

{ "mcpServers": { "handaas-mcp-server": { "type": "http", "url": "http://127.0.0.1:8000/mcp", "args": [] } } }

## Additional configuration options

You may also run the server in STDIO mode for local development, using the complete command shown in the example. This is useful when integrating with Cursor/Cherry Studio MCP configurations.

{ "mcpServers": { "bidding-mcp-server": { "command": "uv", "args": ["run", "mcp", "run", "{workdir}/server/mcp_server.py"], "env": { "PATH": "{workdir}/mcp_env/bin:$PATH", "PYTHONPATH": "{workdir}/mcp_env", "INTEGRATOR_ID": "your_integrator_id", "SECRET_ID": "your_secret_id", "SECRET_KEY": "your_secret_key" } } } }

## Connecting to the official remote service

If you prefer not to run locally, you can connect to the official remote MCP service by configuring a remote URL in your client.

{ "mcpServers": { "bidding-mcp-server":{ "type": "http", "url": "https://mcp.handaas.com/bidding/bidding_bigdata?token={token}" } } }

Note: You must obtain integrator\_id, secret\_id, secret\_key, and token by registering and activating the MCP portal.

## Example questions to try

- Look up companies by fuzzy keywords such as a partial name or brand to discover potential bidders.

- Analyze a company’s bid win statistics to understand its performance across regions and deal sizes.

- Retrieve bidding information for a company to study past announcements, project regions, and involved bidders.

## Security and best practices

Protect your integrator\_id, secret\_id, and secret\_key. Do not expose your credentials in client-side code. Use secure storage for tokens and consider rotating credentials periodically.

## Notes on usage

- When querying for enterprises by full name, you may first search with fuzzy keywords to obtain the exact full name. - Each page of results is limited to 50 items. - Date parameters must use the format YYYY-MM-DD. - Monetary values are expressed in units of ten thousand (万元). - You can filter by province and city using multi-select options.

## Tools and endpoints you can use

The MCP server provides a set of tools that return structured data for analysis. Here are the available tool endpoints and what they do.

## Available tools

### bid\_bigdata\_fuzzy\_search

Performs fuzzy searches for enterprises by keywords such as company names, person names, brands, products, or roles. Returns a list of matching enterprises.

### bid\_bigdata\_bid\_win\_stats

Provides enterprise bid win statistics including item distribution, amount distribution, regional distribution, and bid trends.

### bid\_bigdata\_bidding\_info

Queries bidding information for a specified enterprise, returning announcements, regions, project details, and related bidders.

### bid\_bigdata\_tender\_stats

Provides tender statistics for an enterprise, including item distribution, amount distribution, regional distribution, and tenders trends.

### bid\_bigdata\_procurement\_stats

Offers procurement statistics for an enterprise, including product and regional distributions.

### bid\_bigdata\_bid\_search

Searches bidding information with multiple filters to return detailed bidding announcements.

### bid\_bigdata\_planned\_projects

Queries planned projects announced by an enterprise, returning project details.