AI Agent Hub
Back to skills
⚖️

Conditional Stock Screener

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Install @user_d83d7c97/wm-screen-index according to https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Batch screening is often used for filtering by ROE, PE, dividend yield, or industry tags. The practical difficulty is field semantics and units: mixing real-time and EOD change fields, encoding ratios as integers, or misaligning time windows.

How It Works

This skill is the execution entry point for StockPicker, organized around the following path:
- Align fields first with indicators by name or alias
- Execute batch screening with conditions
- Estimate hit counts with preview_count to avoid overly broad or narrow filters
- Manage personal strategies with list_strategy, run_strategy, and save_strategy

Parameters are usually passed in args, such as market, conditions, top_n, and offset. Results are exposed through rows, row_ids, deeplinks, and freshness, making it easy to read candidate lists or open the workspace.

Boundaries

  • Intended for batch screening, not single-company fundamental cards, watchlists, or index member lookup
  • Ratio fields should use 0.15 or 15%, not 15
  • Use $PERCENT_LAST for same-day change and $CHG_1D_LAST for EOD technical slices
  • After a successful run, display the returned deeplinks as-is rather than constructing result URLs manually

Use Cases

  • When building a weekly stock pool, filter by ROE, PE, and industry tags, preview hit counts, then execute screening.
  • Before changing filters, align ROE, dividend yield, and other fields to correct column names and ratio formats.
  • When reviewing saved strategies, list stored screening strategies and rerun the same condition set.
  • When results are too large, use offset pagination and top_n to limit rows, then save the current filter.

Best For

  • Equity research assistants maintaining stock pools who need batch filters by valuation and ROE.
  • Individual investors managing saved screeners who need to list, rerun, and update filters.
  • Fundamental quant researchers who need correct field names and ratio formats for ROE and dividend yield.
  • Analysts generating candidate lists who need hit-count previews before widening conditions.