← Back to Skills Marketplace
Ac Stock Ultrashort Custom
by
c595390153-a11y
· GitHub ↗
· v1.0.0
· MIT-0
79
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install ac-stock-ultrashort-custom
Description
A股超短线交易策略。用于 Peter 的游牧型超短线交易风格：只做主线、只做最强、1-2天T+1套利。当需要分析A股盘面、筛选热点板块、寻找龙头股、制定买卖策略时使用。
README (SKILL.md)
A股超短线交易策略
核心定位
游牧型超短线交易者 — 只做主线，只做最强，只赚爆发段。
选股标准
1. 识别主线题材
- 政策驱动：重大政策出台、规划发布
- 事件驱动：突发新闻、行业突破
- 资金驱动：板块持续放量、游资/机构大幅买入
- 板块涨停数：≥3只确认题材持续性
2. 筛选条件
| 指标 | 范围 | 说明 |
|---|---|---|
| 流通市值 | 50-300亿 | 弹性与容量兼顾 |
| 换手率 | 5%-15% | 活性适中 |
| 量价配合 | 放量突破关键位置 | 资金认可 |
| 板块涨停数 | ≥3只 | 确认题材持续性 |
3. 选股优先级
龙头 > 跟风，宁可不做也不做杂毛：
- 率先涨停的辨识度最高
- 成交量最大的关注度最高
- 市场讨论最多的情绪最强
- 创业板（300开头）优先 — 20%涨跌停利润更高
- 北交所（8/9开头）可选 — 30%涨幅空间更大
买卖时机
买点模式
| 模式 | 描述 | 适用场景 |
|---|---|---|
| 首板打板 | 题材确认日，龙头首次涨停时排板/打板买入 | 新题材启动第一天 |
| 二板确认 | 首板次日高开5%以上，竞价或盘中确认强度后介入 | 题材已验证，强势确认 |
| 分歧转一致 | 龙头回调后再次放量拉升的转折点 | 题材分歧后重新凝聚 |
卖点规则
| 情况 | 动作 |
|---|---|
| 高开低走跌破均线 | 出局 |
| 封板不牢反复开板 | 减仓或出局 |
| 板块跟风股大面积炸板 | 退潮信号，立刻走 |
| 持股一般不超过3天 | 到点走人 |
仓位管理（50万本金）
| 情况 | 仓位 | 金额 | 标的数 |
|---|---|---|---|
| 强主线 + 强龙头 | 50% | 25万 | 1-2只 |
| 一般机会 | 30% | 15万 | 1只 |
| 弱信号 / 试错 | 10-15% | 5-7万 | 1只 |
铁律：
- 单只股票不超50%仓位
- 永远保留底仓现金（至少5万）
- 绝不临时起意加仓
止损纪律
| 类型 | 规则 | 执行 |
|---|---|---|
| 固定止损 | 买入价 -5% | 无条件执行 |
| 日内止损 | 买入当天收盘亏 >3% | 次日竞价出局 |
| 题材止损 | 主线逻辑被证伪 | 无论盈亏立刻走 |
回撤控制
| 触发条件 | 动作 |
|---|---|
| 单月回撤 ≥10% | 警惕，下一笔交易前强制复盘 |
| 连亏3次 | 强制停手 1-3天，复盘找问题 |
| 账户回撤达15% | 降至半仓以下运作 |
数据来源
每日必看
- 东方财富 - 概念板块资金流向：https://data.eastmoney.com/bkzj/gn.html
- 财联社 - 热点新闻：https://www.cls.cn/hot-stock
- 同花顺/通达信 - 行情软件
复盘流程
盘前（8:30-9:15）
- 复盘昨日涨停板，筛选目标
- 关注集合竞价情况
- 制定当日交易计划
竞价（9:15-9:25）
- 观察目标股竞价强度
- 决定是否参与、仓位多少
盘中
- 只执行计划，不临时起意
- 监控持仓股和板块走势
- 触发止损/卖点立刻执行
盘后（15:00-16:00）
- 复盘当日操作
- 记录交易日志
- 分析主线题材持续性
交易日志模板
日期：2026-XX-XX
标的：
买入价：
卖出价：
盈亏：+X% / -X%
买卖理由：
复盘总结：
核心信条
短线交易的核心不是选股，是纪律。
计划你的交易，交易你的计划。
Usage Guidance
This skill is a text-only trading playbook and is internally consistent with that purpose. Before installing: verify the author's identity/provenance (no homepage provided and _meta.json owner differs from registry metadata), treat the content as opinion not guaranteed performance, do not grant any trading account credentials or automated trading permissions to the skill, and backtest or paper-trade the rules before using real capital. If you plan to let an agent act autonomously with trading accounts, add explicit safeguards (manual approval, rate/size limits, and separate, minimal-permission API keys).
Capability Analysis
Type: OpenClaw Skill
Name: ac-stock-ultrashort-custom
Version: 1.0.0
The skill bundle contains only documentation and strategy instructions for an AI agent to act as an A-share stock trading assistant. There is no executable code, no data exfiltration logic, and no malicious prompt injection; it simply outlines trading criteria, risk management rules, and legitimate financial data sources (e.g., Eastmoney, CLS).
Capability Assessment
Purpose & Capability
The name and description describe an A-share ultra-short trading strategy and the SKILL.md contains only strategy rules, sources, and templates — nothing else is requested. Minor metadata inconsistency: _meta.json ownerId/slug differ from the registry metadata and there is no homepage/source link; this affects provenance but not functional coherence.
Instruction Scope
Runtime instructions are purely procedural trading guidance (selection criteria, buy/sell rules, risk and log templates). They reference public market data sites for research but do not instruct reading local files, accessing secrets, or sending data to unknown endpoints.
Install Mechanism
No install specification and no code files — the skill is instruction-only, which minimizes filesystem and runtime risk.
Credentials
The skill requires no environment variables, credentials, or config paths. The external URLs listed are public data sources relevant to the strategy.
Persistence & Privilege
Default invocation settings (not always, agent-invocable allowed) are present. There is no request for elevated or persistent system privileges and the skill does not attempt to modify other skills or global config.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ac-stock-ultrashort-custom - After installation, invoke the skill by name or use
/ac-stock-ultrashort-custom - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of ac-stock-ultrashort-custom skill, focused on A股主线龙头超短线交易策略
- 明确买入、卖出、止损、仓位及回撤管理等全流程纪律，适配游牧型超短风格
- 提供主线题材筛选及龙头股挑选标准，包含具体量化指标
- 覆盖每日复盘、盘前计划、盘中执行及交易日志模板
- 推荐实时数据与资讯来源（东方财富、财联社等）
Metadata
Frequently Asked Questions
What is Ac Stock Ultrashort Custom?
A股超短线交易策略。用于 Peter 的游牧型超短线交易风格：只做主线、只做最强、1-2天T+1套利。当需要分析A股盘面、筛选热点板块、寻找龙头股、制定买卖策略时使用。 It is an AI Agent Skill for Claude Code / OpenClaw, with 79 downloads so far.
How do I install Ac Stock Ultrashort Custom?
Run "/install ac-stock-ultrashort-custom" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Ac Stock Ultrashort Custom free?
Yes, Ac Stock Ultrashort Custom is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Ac Stock Ultrashort Custom support?
Ac Stock Ultrashort Custom is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Ac Stock Ultrashort Custom?
It is built and maintained by c595390153-a11y (@c595390153-a11y); the current version is v1.0.0.
More Skills