← 返回 Skills 市场
kobenfang

BigA

作者 kobenfang · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ suspicious
89
总下载
0
收藏
0
当前安装
6
版本数
在 OpenClaw 中安装
/install biga
功能描述
A股智能分析与选股工具。维护动态股票池(最多30支),按高科技×中小市值×好业绩原则筛选,推送买卖信号。含独立技术面择时分(-10~+10)用于判断买入卖出时机。
使用说明 (SKILL.md)

📈 BigA · A股智能选股

核心能力

  1. 股票池管理:动态维护≤30支,超额自动淘汰
  2. 实时行情:腾讯API http://qt.gtimg.cn/q=\x3C前缀+代码>(批量逗号分隔,最多900)
  3. 评级:🔥买入(≥70)/👀持有(50-70)/⚠️卖出(\x3C40)/📌观察
  4. 选股:高科技(半导体/AI/新能源/光模块/液冷)× 中小市值(50-500亿)× 好业绩
  5. 定时推送:早8:30开盘前瞻/盘中9:30-14:30扫描/15:30收盘复盘

推送格式

所有输出以 📈 BigA · 大A 开头, 💬 回复"股票池"查看 结尾。

时间窗口

场景 窗口
开盘前瞻(08:30) 过去12h(昨晚20:00→今早8:30)
盘中扫描 过去1h(仅9:30-15:00)
收盘复盘(15:30) 当日全天
用户触发 按要求,默认当天

数据源

行情(腾讯)

curl -s "http://qt.gtimg.cn/q=sh688313" → ~分隔字段 核心: 3=价格, 32=涨跌幅%, 38=换手率%, 39=PE, 44=振幅%, 46=总市值

K线(Sina主/腾讯备)

主:http://money.finance.sina.com.cn/...getKLineData?symbol=\x3C前缀+代码>&scale=240&ma=5&datalen=60 备:http://web.svip.ifzq.gtimg.cn/appstock/app/fqkline/get?param=\x3C前缀+代码>,day,,,1 前缀: 沪sh/深sz/北交bj 返回JSON: day, open, high, low, close, volume, ma_price5, ma_volume5

股票池评分(BigA, 0-100)

维度 权重 评分
基本面 40 PE合理(15-40,15分)+营收增速>20%(15分)+净利增(10分)
催化剂 25 政策利好(10)+产业向上(10)+消息(5)
技术面 20 趋势向上(10)+量价(5)+相对强弱(5)
热度 15 板块热度(8)+资金流入(7)

管理: 初始60📌 | >30支淘汰最低分 | 连3次\x3C40自动移除

技术面择时分(独立体系, -10~+10)

与BigA评分正交:BigA回答"股票好不好",择时分回答"何时买卖" 完整细则见references/technical-timing-score.md

维度 评分规则 数据
动量(±2) 近5日:\x3C-15%→+2, +15~25%→-2, >+25%→-2 K线
日内量价(±2) 低开高走+振幅>5%→+2, 高开低走+放量→-2 当日K线
成交量(±2) 地量(\x3C均值0.5)→+1, 天量(>均值3)→-2 换手率
相对强度(±2) 独立大跌→+2, 跟风→-2 个股vs指数
极端修正(±2) 60日新低→+1, 60日新高→-1, 连续3板开板→±2 —

总分判定: ≥+6🔥买入 / +3~+5✅买入 / -2~+2👀持有 / -5~-3⚠️仅持有 / ≤-6🛑止盈

选股硬性规则

入池:高科技×50-500亿×营收>20%增长×有催化剂 排除:传统行业/ST/PE>100无高增速/一周涨>30%

工作流程

安装检测

每次触发检查memory/biga-installed,不存在则提示安装。 用户说「帮我安装」→ cron add 3个job(templates见references/):

  • biga-morning: 30 8 * * 1-5(180s)
  • biga-scan: 30 9,10,11,13,14 * * 1-5(180s)
  • biga-evening: 30 15 * * 1-5(240s)

收盘复盘流程(15:30)

  1. 读memory/biga-stock-pool.md + references/user-preferences.md
  2. 腾讯API获取行情-所有池中股票+大盘指数
  3. Sina拉日K线(60天),存memory/biga-technical-data.md(增量更新)
  4. 计算BigA评分(基本面对照财务数据+催化剂+技术面+热度)
  5. 计算技术择时分(4维度+修正),综合判断
  6. 去重(24h同方向不重复),查热点
  7. 输出:只推买入/卖出标的,标的格式加择时:+X ✅字段

开盘前瞻流程(08:30)

1-2同上 → 搜索隔夜外围 → 输出买入信号(不加技术分)

盘中扫描(仅异动时推)

池中涨跌幅>5%或成交量>50%才推

输出约束

  • 推送≤800字(开盘/复盘),≤200字(扫描)

  • 3条买卖信号只展Top3

  • 无信号→空输出,一字不发
  • 节假日休市不执行;开盘前瞻改外围追踪

常见问答

Q: 实时行情? → exec+curl腾讯API Q: 财务数据? → web_search "\x3C代码> 财报" Q: 重复推送? → 24h同方向不重复 Q: 入池/删股? → "关注/删除 \x3C名称>" Q: 节假日? → 休市不执行 Q: 评分? → 基本面40+催化剂25+技术面20+热度15

文件索引

运行时(memory/)

  • biga-stock-pool.md - 股票池(≤30支)
  • biga-sent-signals.md - 24h推送去重
  • biga-retry-queue.md - 失败重试
  • biga-technical-data.md - K线缓存
  • biga-installed - 安装标记

参考(references/)

  • cron-templates.json - Cron job模板
  • cron-install-shell.sh - 安装脚本
  • technical-timing-score.md - 技术评分细则
  • user-preferences.md - 用户偏好
  • sector-matrix.md - 板块分析
安全使用建议
Before installing, read the cron shell template and only run it if you want recurring weekday stock analysis. Know how to remove the BigA cron jobs, periodically review the stored stock-pool/preference memory, and verify any buy or sell signal independently before acting on it.
功能分析
Type: OpenClaw Skill Name: biga Version: 1.1.0 The skill implements an automated stock analysis tool that utilizes high-risk capabilities, specifically shell execution via `exec+curl` to fetch data from external financial APIs (Tencent/Sina) and persistence through the `openclaw cron` scheduling system. While these behaviors are aligned with the stated purpose of providing periodic stock market updates, the use of direct shell commands and automated task scheduling represents a significant attack surface. No clear evidence of intentional malice or data exfiltration was found in the analyzed files (SKILL.md, cron-install-shell.sh).
能力评估
ℹ Purpose & Capability
The stock screening, timing score, stock-pool maintenance, and alerting behavior match the stated A-share analysis purpose. The artifacts do not show trading-account access or order execution, but the outputs are framed as buy/sell signals.
ℹ Instruction Scope
The workflows are scoped to market data, searches, scoring, and concise alerts. Scheduled prompts also require direct or empty output, which is purpose-aligned for alerts but can omit context unless the user asks.
ℹ Install Mechanism
Registry metadata says there is no install spec, while included reference files provide manual openclaw cron commands for three weekday jobs. This is disclosed, but users should review and understand those commands before running them.
ℹ Credentials
No credentials or required config paths are declared. The skill expects exec+curl, web_search, and HTTP market-data APIs, which are proportionate to stock analysis but send requested stock symbols and queries to external providers.
ℹ Persistence & Privilege
It keeps runtime memory files for the stock pool, signal deduplication, retries, and K-line cache, and it can install recurring jobs after user approval. This is bounded to the skill’s purpose but should be reviewable and removable by the user.
如何使用
  1. 确保已安装 OpenClaw（本地或 Docker 部署）
  2. 在对话框中输入安装命令：/install biga
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /biga 触发
  4. 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.1.0
新增技术面择时分(-10~+10, 4维度+决策矩阵); K线数据源切Sina主/腾讯备; 精简SKILL.md
v1.0.4
Step 0 新增 memory/biga-skill-detected 标记创建，供 Eyes 判断是否推广
v1.0.3
新增触发词「装大A」「安装大A」
v1.0.2
安装流程改用 cron 工具注册，不再 exec 执行脚本
v1.0.1
补充盘中扫描有效时段 9:30~15:00，与 Eyes 保持一致
v1.0.0
初版发布：A股智能股票池+腾讯API实时行情+买卖信号推送
元数据
Slug biga
版本 1.1.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 6
常见问题

BigA 是什么？

A股智能分析与选股工具。维护动态股票池(最多30支),按高科技×中小市值×好业绩原则筛选,推送买卖信号。含独立技术面择时分(-10~+10)用于判断买入卖出时机。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 89 次。

如何安装 BigA？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install biga」即可一键安装，无需额外配置。

BigA 是免费的吗？

是的，BigA 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

BigA 支持哪些平台？

BigA 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 BigA？

由 kobenfang（@kobenfang）开发并维护，当前版本 v1.1.0。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论