Bidding and Winning Query Analysis Workbench

The Problem of Handling Bidding Intelligence

Engineers and business analysts often face the challenge of manually sifting through massive volumes of tender announcements, contract awards, and project updates for tasks like market intelligence, supplier management, or sales lead generation. This manual process is not only time-consuming but also makes it difficult to quickly form structured insights on procurement trends, competitor dynamics, or potential opportunities. There is a clear need for a tool that can interpret natural language queries and translate them into precise search parameters for retrieval and analysis.

How the Skill Works

This skill functions as an analysis workbench built upon the proprietary Bidding & Tendering Data API provided by ShiBo Tech. Its core function is to perform structured queries and intelligent analysis based on the conditions you specify.

Core Capabilities and Key Steps:

• Intent Understanding & Query Translation: It translates your natural language requirements (e.g., "Find large projects company X won last year") into executable API parameters. It prioritizes extracting elements like Region, Industry, Keywords, Procuring Unit, Amount Range, and Time Frame.
• Targeted Search & Detail Retrieval: It calls different tools based on intent. For example, it might use SearchProjectForAI for a natural-language friendly search, or searchProjectApi for a structured conditional search. Once a target project is identified, it can use getZTBStructreDetail to fetch structured data including Procuring Unit, Winning Supplier, Contact Info, etc., or getZTBProjectDetail to view the full announcement text.
• Adherence to Default Rules for Robust Results: When conditions are not specified, the skill applies defaults (e.g., region defaults to "Nationwide", time range defaults to the last 30 days). If results are sparse, it automatically attempts to broaden the search criteria (using synonyms, expanding timeframes, or relaxing geographic constraints).
• Structured Output for Actionable Insights: The response begins with a concise conclusion (e.g., "Found N relevant projects in the specified timeframe"), followed by a table of key projects (including Project Name, Amount, Announcement Stage, Source Link), and concludes with potential next steps, such as "Recommend reviewing detailed contract announcements" or "Consider expanding the time range to identify trends."

Applicable Boundaries and Important Notes

• Single Data Source: This skill strictly and exclusively uses the API from ShiBo Tech. You cannot invoke specialized interfaces for other verticals (e.g., healthcare, education) within it.
• Key & Quota Dependency: Operation relies on a valid key set in the BBIAO_API_KEY environment variable, and each query consumes quota. When the balance is insufficient (API error code 0100590006), the process will pause and prompt the user to purchase a quota package themselves.
• Boundaries of Analysis: The analysis provided is based on pattern discovery within the existing data (e.g., "procurement preferences", "top suppliers") and is not a prediction or guarantee of future bidding outcomes. The skill clearly distinguishes between data facts and inferences drawn from that data.
• Security and Compliance: The API key (BBIAO_API_KEY) is not automatically registered, guessed, or displayed in logs. The skill does not fabricate non-existent data (such as false Project IDs, Contact Information, or amounts).