rhzy-contractor-due-diligence

Collects contractor qualification, litigation, and penalty data from six Chinese government sources into structured JSON.

Updated Sep 2, 2026
One-click install
npx skills add https://github.com/pinedu/xmc-work-wiki --skill rhzy-contractor-due-diligence-pinedu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rhzy-contractor-due-diligence
Source: https://github.com/pinedu/xmc-work-wiki/tree/main/skills/rhzy-contractor-due-diligence
Command: npx skills add https://github.com/pinedu/xmc-work-wiki --skill rhzy-contractor-due-diligence-pinedu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, lxml, playwright, and includes references (resource) components.

What problem does it solve? Before signing or renewing a contract with a construction general contractor, you need verifiable facts about their qualifications, lawsuits, penalties, and credit status scattered across multiple Chinese government websites. This Skill automates that due diligence data collection and outputs clean JSON ready for risk assessment. ## Core Features & Use Cases - Six-source collection: Queries the National Enterprise Credit Information System, China Enforcement Information Disclosure, China Judgments Online, MOHURD's construction qualification platform, Guizhou Provincial Housing Department, and the State Taxation Administration's major tax violation list. - Fault-tolerant pipeline: Each source fails independently without blocking others; missing dimensions are flagged in the merged JSON for downstream weight adjustment. - Offline mode: When the network is unavailable, generates placeholder JSON with a TODO checklist so users can manually fill in data from browser lookups. - Use Case: Before negotiating with a new general contractor in Guizhou, run the collector with the company name and credit code to produce a merged JSON file, then feed it to the rhzy-skill1 risk assessment layer for a four-level risk rating. ## Quick Start Ask the agent to run due diligence on a construction company by providing its full company name and unified social credit code, for example: run due diligence on Guizhou Construction Group First Building Engineering Co., Ltd.

Frequently Asked Questions about rhzy-contractor-due-diligence

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run due diligence on a Chinese construction company?

Run the collect.py script with the company name and unified social credit code as arguments. It queries six government sources serially, writes one JSON per source, and merges them into a single file for downstream risk assessment.

What data sources are used for contractor background checks in China?

This Skill uses six public sources: the National Enterprise Credit Information System, China Enforcement Information Disclosure, China Judgments Online, MOHURD's construction qualification platform, the Guizhou Housing Department, and the national major tax violation list.

Does the collection script work without internet access?

Yes, it includes an offline mode. When network checks fail, the script generates placeholder JSON files with TODO checklists so you can manually look up each source in a browser and fill in the results.

Why does the gsxt.gov.cn source get skipped during collection?

The National Enterprise Credit Information System requires a login session. You must manually log in, export browser cookies to ~/.hermes/cookies/gsxt.json, and the session lasts about 30 minutes. Without that file, the source is skipped.

What happens when one data source fails during collection?

Each source runs in an independent try/except block, so one failure does not block the others. Failed sources are recorded in the merged JSON's collection_meta, and the downstream risk assessment reweights the remaining dimensions.