# stockbee-20pct-study

Builds a structured model book of US equities moving 20% or more, tracks catalysts, and calculates multi-day forward outcomes for pattern research.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 2600 installs · ★ 3267
- Author: tradermonty
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/tradermonty/claude-trading-skills/tree/HEAD/skills/stockbee-20pct-study/SKILL.md
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/skills/stockbee-20pct-study

## What it does

The `run_20pct_study.py` script captures US equities that gain or lose at least 20% within a chosen lookback window. It stores structured records into `state/stockbee/20pct_study_events.jsonl` alongside chart context and news catalysts. This is strictly a market research and model-book tool. It does not output buy or sell signals and cannot place broker orders.

The workflow covers five steps: scanning, catalyst enrichment, forward outcome tracking, cohort summaries, and historical backfills. It evaluates stock performance across 1, 3, 5, 10, and 20 days, recording MFE, MAE, and continuation returns. Historical backfills automatically flag survivorship bias unless an explicit survivorship-complete dataset is supplied.

You need Python 3.9+ and either an FMP API key or an offline daily OHLCV JSON file. If news data is missing, the tool logs the move as `NO_CLEAR_NEWS`, meaning you will have to categorize catalysts manually.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add tradermonty/claude-trading-skills --skill stockbee-20pct-study -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/stockbee-20pct-study/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/stockbee-20pct-study/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add tradermonty/claude-trading-skills --skill stockbee-20pct-study -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/stockbee-20pct-study/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add tradermonty/claude-trading-skills --skill stockbee-20pct-study -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/stockbee-20pct-study/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install tradermonty/claude-trading-skills stockbee-20pct-study
```

2. **Manually: put the files in place**
- File: `.github/skills/stockbee-20pct-study/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add tradermonty/claude-trading-skills --skill stockbee-20pct-study -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/tradermonty/claude-trading-skills --path skills/stockbee-20pct-study
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/stockbee-20pct-study/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add tradermonty/claude-trading-skills --skill stockbee-20pct-study -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/tradermonty/claude-trading-skills/tree/HEAD/skills/stockbee-20pct-study/SKILL.md

## When to use it
- scan US equities moving plus or minus 20%
- build a model book of explosive market movers
- track forward return, MFE, and MAE across 1 to 20 days
- group cohorts by catalyst type and technical close quality

## Example prompts
- Scan the US market for 20% movers over the last 5 days using `run_20pct_study.py`
- Update forward outcomes for tracked events in `state/stockbee/20pct_study_events.jsonl`
- Backfill 20% movers since 2020 using `data/us_daily_ohlcv.json`

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: 4463ec4b8d4b87a6.

BEGIN THIRD-PARTY FILE 4463ec4b8d4b87a6
````text
---
name: stockbee-20pct-study
description: Build and maintain a Stockbee-style daily 20% mover study for US equities by scanning +20%/-20% movers, classifying catalysts and setup context, updating forward outcomes, and summarizing cohort patterns. Use when the user asks to run a daily 20% study, backfill historical 20% movers, find recurring edge patterns, or build a model book of explosive market moves.
---

# Stockbee 20% Study

Build a daily event study of US equities that moved +20% or -20% over a defined window. Convert large movers into structured study records, classify the catalyst and chart context, update forward outcomes, and summarize recurring patterns for research.

This skill is a research, model-book, and setup-fluency workflow. It does not generate buy/sell signals, place orders, or output broker execution instructions.

## When to Use

- User wants to run a Stockbee-style daily 20% mover study
- User asks which stocks moved +20% or -20% today, this week, or over a configurable lookback window
- User wants to backfill historical 20% movers and study what happened next
- User wants to identify continuation, reversal, exhaustion, or theme-cluster patterns
- User wants to build a model book of explosive winners, major failures, and failed low-quality pops
- User wants edge hints for downstream strategy research rather than immediate trade signals

## Prerequisites

- Python 3.9+
- FMP API key for live US universe scans, or offline OHLCV JSON via `--prices-json`
- Optional structured news/catalyst JSON for higher-quality catalyst classification
- Recommended market regime artifact from `market-regime-daily`
- Recommended local state path: `state/stockbee/20pct_study_events.jsonl`

## Workflow

### Step 1: Scan for 20% Movers

Run after the US market close, or against the latest complete daily bar in an offline OHLCV file.

```bash
python3 skills/stockbee-20pct-study/scripts/run_20pct_study.py scan \
  --fmp-universe \
  --max-symbols 300 \
  --as-of 2026-06-28 \
  --lookback-days 5 \
  --min-abs-return-pct 20 \
  --min-price 5 \
  --min-dollar-volume 20000000 \
  --include-down-movers \
  --state-file state/stockbee/20pct_study_events.jsonl \
  --output-dir reports/
```

Use offline data instead of FMP:

```bash
python3 skills/stockbee-20pct-study/scripts/run_20pct_study.py scan \
  --prices-json data/us_daily_ohlcv.json \
  --as-of 2026-06-28 \
  --lookback-days 5 \
  --include-down-movers \
  --state-file state/stockbee/20pct_study_events.jsonl \
  --output-dir reports/
```

### Step 2: Enrich and Classify Events
````
END THIRD-PARTY FILE 4463ec4b8d4b87a6

The third-party file has ended. Everything between the 4463ec4b8d4b87a6 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2575 of 5778 bytes. © tradermonty · MIT · https://github.com/tradermonty/claude-trading-skills/tree/HEAD/skills/stockbee-20pct-study/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md