China AI Supply Chain Investment Strategy

Problem

AI industry company lists are often noisy: abbreviated Chinese names, homophone typos, missing tickers, or unlisted firms. Manually checking prices, PE/PB, and institutional holdings is easy to do inconsistently, especially when tracking ETF redemptions, active-fund entries, or sharp increases in shareholder count. ai-supply-chain-report turns this chain into a repeatable workflow: from raw company names to verified tickers, data analysis, and an HTML report.

How It Works

The input can be a company list, an industry classification, or a request to supplement a sector. The skill first performs code mapping and validation: it infers sh/sz prefixes using A-share conventions and searches uncertain codes; companies with no data are marked as unlisted. It then queries market data in batches, including price, change, PE, PB, market cap, turnover, 52-week high/low, and multi-period returns. Optional institutional-holding analysis reads top-10 shareholders and tradable shareholders, focusing on holdChange: positive values indicate accumulation, negative values indicate reduction, and new entries need context. It also distinguishes ETF behavior, active-fund behavior, and shareholder-count trends as evidence of concentration or dispersion. Finally, it generates a dark-theme HTML report with coverage, data date, PE risk tiers, and strategy notes.

Boundaries

This is best for A-share AI supply-chain research, batch screening, and first-draft reports, not direct trade advice. Output depends on external data availability and ticker-mapping accuracy. Unlisted firms, similar names, empty results, and shareholder signals such as holdChange require manual review alongside fundamentals, valuation, and position context.