Longxia Stock Market Analysis
Paste the following prompt into your AI chat to install this skill:
Please install @user_47b3804c/stock-analysis-lxx using the guide at https://skillhub.cn/install/skillhub.md.
About this skill
Problem
Equity research often fails not because of missing models, but because the workflow is scattered: reading news first, patching together financials, valuation, and announcements, then turning candidates into company encyclopedias. Longxia Stock Market Analysis keeps the task anchored to a local data layer: scanning, ranking, and candidate screening run through scripts first, reading data/output/stock_dataset.csv and data/output/manual_review_queue.csv before deciding what is missing. It breaks low-valuation, high-growth analysis into valuation cheapness, long-term growth, realization quality, catalysts, and risk adjustment, mapped to valuation score, growth score, catalyst score, risk score, and a composite score.
Workflow
The skill operates by task mode: low-valuation high-growth screening, candidate ranking, single-stock analysis, policy mapping, and risk screening. If the user does not specify, it defaults to screening. For quick screening, it sorts by composite score first, then manually checks valuation labels, growth labels, data completeness, and conclusion confidence for the top 5-10 names; deeper explanation is limited to the top 3-5 names by default. If manual_review_queue.csv is non-empty, it fills only the missing fields, writes them back to data/manual_overrides.csv, and reruns the script, avoiding escalation from one missing field to a full manual review. Output should lead with a table, then conclusions, compressing numbers into judgments: current valuation, valuation assessment, long-term growth engine, core catalyst, core contradiction, and a verdict of priority / track / watch / avoid.
Boundaries
It is intended for research reference and does not replace personalized investment advice. The seed universe is not a hard boundary; users can request broader market candidates, but manual overrides must still follow the data-layer constraints. If data completeness or conclusion confidence is low, the skill should explicitly downgrade the conclusion and separate facts from inferences.
Use Cases
- Scan an A-share candidate pool for low valuation and high growth, rank by composite score, and tag priority, tracking, watching, or avoidance.
- Compress one stock into valuation, growth engine, core catalyst, core contradiction, and conclusion confidence.
- Fill only missing fields manually, write the overrides back, rerun the script, and output candidate conclusions.
- After a policy theme appears, map related candidates to catalyst and risk dimensions and produce tracking conclusions.
Best For
- Independent equity researchers who need to compress candidate pools into low-valuation, high-growth rankings.
- Buy-side or sell-side analysts who need single-stock analysis separating facts, inferences, and confidence levels.
- Research assistants responsible for filling missing fields, rerunning scripts, and updating candidate labels.
- Investment operations staff who turn announcement catalysts and risks into priority, tracking, watching, or avoidance tags.
Related Skills
Switch AI from a polite executor into a skeptical thinking partner, using assumption checks, pre-mortems, and evidence gates to expose weak ideas early.
Applies research discipline, data lookup, evidence grading, red-team checks, and structured reporting to analyze AI company investment logic.
Uses Qichacha or Tianyancha MCP data and public information to score corporate credit across 12 dimensions and output a Markdown risk report.
Calculates personal injury compensation via CLI and outputs case summary, standards, line items, sources, and risk notes.