Persistent Nationwide Bidding Data Query

Problem

In bidding workflows, users often need to check project announcements, winning suppliers, a company’s business focus, partnership signals, and market trends for top purchasers, suppliers, brands, and price history. Manual searching can be inconsistent and makes it hard to connect companies, bids, and market data into one analytical path.

How it works

The skill wraps Zhiliao Bidding’s query capabilities into 16 tools, including search_bids, get_bid_detail, search_company, find_competitors, get_top_suppliers, and get_price_trends. A user can ask in natural language, and the agent selects the appropriate tool and builds the request. The main flow is: verify that the weixinpay plugin is available; read or create a stable local X-Device-Id; call /api_pay/{tool}; if the response is 402, trigger WeChat Pay for a usage pack and retry the original request after payment. Credits are bound to the device, so the same balance can be reused across sessions and days, which is useful for ongoing tracking of the same customers, projects, and competitors.

Boundaries

This version depends on the WeChat Pay plugin and a local device ID, so it may not work in environments that cannot complete payment. On failure, it should state the limitation honestly rather than invent bid data or steer users to other sources. It returns API-based query results; offline caching, batch export, or custom BI dashboards still require external systems.