菜单
A股量化选股扫描器，模型驱动自适应因子评分。通过 Tushare 扫描全 A 股 5000+ 只股票，检测市场环境（波动率、动量、资金面、板块轮动），动态调整因子权重，输出选股榜/异动榜/行业分布的 HTML 报告。适用于 A 股选股、量化筛选、因子分析、市场环境诊断、板块轮动、动量策略、价值投资筛选。Use for Chinese A-share stock screening, quantitative stock picking, sector rotation, market regime detection.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
3-layer personality system for AI agents: Yuan (base persona), Ishiki (private self-awareness), Public Ishiki (external behavior rules). Give your agent a consistent character, emotional depth, and safe guest interaction boundaries. Use when: (1) Want agent to have a consistent personality, (2) Agent needs different behavior for owner vs strangers, (3) Building a personal AI companion, (4) Agent responses feel generic and robotic, (5) User says 'be more like...', 'act as...', or wants to customize agent character.
Auto-snapshot workspace before dangerous commands (rm -rf, git clean, git reset --hard). Zero-cost hardlink dedup on macOS/Linux, robocopy mirror on Windows. Use when: (1) About to run destructive file operations, (2) User asks to protect workspace files, (3) Before bulk delete/move operations, (4) User mentions fear of losing files. Also provides snapshot listing and one-command restore.
Click-to-zoom lightbox for chat images, screenshots, and generated visuals. Scroll-wheel zoom (0.5x–5x), drag-to-pan, pinch gesture on mobile, keyboard shortcuts. Zero dependencies — pure CSS + vanilla JS. Use when: (1) Chat contains images the user wants to inspect, (2) Browser screenshots need closer examination, (3) Generated images are too small in the chat bubble, (4) User says 'zoom in' or 'enlarge' about an image.
Persistent cross-session memory system with fact extraction, semantic search, and compiled memory layers. Your agent remembers everything across sessions — user preferences, project context, past decisions, personal details. Use when: (1) User references something from a previous conversation, (2) Agent needs project context it learned before, (3) User says 'remember this' or 'do you remember when...', (4) Starting a new session that should inherit past knowledge, (5) Agent keeps forgetting things between sessions.
Quantitative stock scanner with model-driven adaptive factor scoring. Scans ~100 US large/mid-cap stocks via Finnhub API, detects market regime (volatility, momentum, breadth), dynamically adjusts factor weights, and produces ranked selection/anomaly lists with HTML reports. Use for US stock screening, quantitative stock picking, factor analysis, market regime detection, stock scoring, building a model portfolio, or finding momentum/value/quality plays.
Route different task types to the best-fit model automatically. Chat → fast small model, coding → strong reasoning model, images → vision model, long documents → large context model. Use when: (1) Agent handles mixed tasks (chat + code + vision), (2) Want to optimize speed and cost by matching task to model, (3) User sends an image but current model is text-only, (4) User asks for code review or complex reasoning, (5) User pastes long documents exceeding small model context.
基于 SOC 职业分类
| name | a-share-scanner |
| description | A股量化选股扫描器，模型驱动自适应因子评分。通过 Tushare 扫描全 A 股 5000+ 只股票，检测市场环境（波动率、动量、资金面、板块轮动），动态调整因子权重，输出选股榜/异动榜/行业分布的 HTML 报告。适用于 A 股选股、量化筛选、因子分析、市场环境诊断、板块轮动、动量策略、价值投资筛选。Use for Chinese A-share stock screening, quantitative stock picking, sector rotation, market regime detection. |
| version | 1.0.0 |
| homepage | https://tushare.pro |
| commands | ["/a_scan - 运行全A股量化扫描，生成HTML报告","/a_top - 快速查看今日Top10选股","/a_regime - 查看当前市场环境诊断","/a_sector - 查看板块轮动与行业分布","/a_analyze - 深度分析指定股票（如 /a_analyze 000001 600519）"] |
| metadata | {"clawdbot":{"emoji":"🇨🇳","requires":{"bins":["python3"],"env":["TUSHARE_TOKEN"]},"install":[]}} |
模型驱动的全 A 股量化选股工具。扫描 5000+ 只 A 股，检测市场环境，自适应调整 6 因子权重，输出选股/异动排名 + HTML 报告。
必需： 一个免费的 Tushare Pro token。
export TUSHARE_TOKEN="your_token_here"
或直接传参：
python3 {baseDir}/scripts/a_share_scan.py --token YOUR_TOKEN
依赖安装（自动）：
pip install tushare pandas
或用 uv（推荐）：
uv run {baseDir}/scripts/a_share_scan.py
python3 {baseDir}/scripts/a_share_scan.py
执行流程：
python3 {baseDir}/scripts/a_share_scan.py --top 10
python3 {baseDir}/scripts/a_share_scan.py --codes 000001,600519,300750
python3 {baseDir}/scripts/a_share_scan.py --format json
每只股票按 6 个因子评分，权重根据市场环境动态调整：
| 因子 | 基础权重 | 趋势市调整 | 震荡市调整 | 高波动调整 |
|---|---|---|---|---|
| 动量（当日涨跌+5日+20日涨幅） | 25% | x1.3 加强 | x0.7 减弱 | — |
| 量能（量比+换手率+成交额） | 20% | — | — | — |
| 估值（PE/PB/市值合理性） | 15% | x0.8 降低 | x1.3 提升 | — |
| 成长（营收增速+净利增速） | 20% | — | — | x1.2 提升 |
| 分析师（评级+目标价空间） | 10% | — | — | — |
| 风险调整（波动率+涨停风险+偏离度） | 10% | — | — | x1.4 加重 |
扫描器分析全市场数据，自动判断：
| 接口 | 数据 |
|---|---|
daily | 全A股日线行情（OHLCV+涨跌幅+换手率） |
daily_basic | 量比、总市值、流通市值、PE、PB |
stock_basic | 股票名称、行业、上市日期 |
trade_cal | 交易日历 |
全 A 股（上交所 + 深交所 + 北交所），自动过滤：
精美响应式 HTML，包含：
完整机器可读输出，含全部评分数据。
用户：帮我扫一下今天A股有什么好票
→ /a_scan
用户：现在市场什么环境？
→ /a_regime
用户：看看半导体和新能源板块
→ /a_sector
用户：分析一下宁德时代和比亚迪
→ /a_analyze 300750 002594
用户：给我一个快速Top10
→ /a_top
Part of Lynn — a personal AI agent with memory and soul. Lynn has this built-in by default.
Lynn 桌面智能体内置了 A 股扫描器：
git clone https://github.com/LynnMerkyor/Lynn.git
cd Lynn && npm install
# 然后直接聊天：
# "扫一下A股"
# "分析一下茅台和宁德"
# "设置每天收盘后自动扫描"