qiuqiubuchongle-cloud/chokepoint-atlas

This is a research tool for studying US stocks in the AI industry, with a focus on supply chain bottlenecks. The idea behind it is different from typical stock screeners or AI chat tools. Instead of asking which stock is worth buying right now, it asks: if demand in a sector keeps growing, which part of the supply chain will run out of capacity first? That bottleneck is where the most constrained companies tend to sit, and constrained companies in high-demand industries often have pricing power. The workflow starts by picking a specific real-world system to study, such as an AI data center, a robot actuator chain, or advanced chip packaging infrastructure. The tool then breaks that system into layers, from end-customer demand down through system integration, core components, materials, and upstream tooling. It identifies where the system is most likely to get stuck, then pulls evidence from earnings reports, investor calls, company websites, news, and analyst research to back up the analysis. There are three main ways to use it. The first is for a single research focus you have already chosen: you run a script that produces a set of structured output files including a quick scan, an evidence memo, a scoring card, and a catalyst watch list. The second is for comparing multiple research directions side by side so you can prioritize where to look first. The third is for feeding in raw materials you have already collected, such as clippings or notes, and having the tool organize them into the same structured format. The README is written in Chinese, with an English product description available in a separate file. The tool is positioned as a structured research process aid rather than a recommendation engine. It does not tell you what to buy but helps you build a traceable, reusable evidence base before drawing conclusions. The README is sparse on technical setup details, pointing instead to a product manual and skill reference for deeper guidance.