Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

股票投资智投顾问

v1.0.0

提供沪深北港美美股实时行情、多维数据分析与专业投资报告的全流程股票综合智投顾问服务。

0· 91·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for gouyujun/stock-investment-advisor.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "股票投资智投顾问" (gouyujun/stock-investment-advisor) from ClawHub.
Skill page: https://clawhub.ai/gouyujun/stock-investment-advisor
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install stock-investment-advisor

ClawHub CLI

Package manager switcher

npx clawhub@latest install stock-investment-advisor
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The skill claims to integrate A‑share, HK, US real‑time data, K‑line extraction and Feishu document generation, but the repository only contains a placeholder scripts/analyze.py; SKILL.md and README reference multiple sub-scripts (scripts/a-stock/, scripts/glmv/, scripts/autoglm/) and a venv path that are not present. Generating Feishu documents implies Feishu API credentials, yet no required env vars or primary credential are declared. This mismatch suggests the published bundle is incomplete or expects external components not included.
!
Instruction Scope
Runtime instructions direct the agent to run local scripts (uv run {SKILL_DIR}/scripts/a-stock/analyze.py, {SKILL_DIR}/scripts/glmv/fetch_all.py), use tools like web_search, image, read, and feishu_create_doc, and to 'must read' generated images. Those commands assume presence of additional code and a Feishu integration. The provided analyze.py is a harmless placeholder that does not perform network data collection, so the instructions and actual runnable code diverge. The instructions also instruct writing and reading files under the user's home workspace (~/.openclaw/workspace), which is within agent filesystem scope but should be noted.
Install Mechanism
No install spec is provided (instruction-only plus one script), which limits automatic risk from arbitrary installers. However SKILL.md expects Python venvs and uv-managed scripts in subfolders and refers to running a specific interpreter in a venv path that does not exist in the bundle. This is not an active install risk but indicates missing components or implicit install steps that were not provided.
!
Credentials
The README and SKILL.md mention optional TUSHARE_TOKEN and Feishu API usage, yet requires.env is empty and no primary credential is declared. Creating cloud docs via Feishu will require credentials; the skill offers no explicit declaration of which env vars, scopes, or tokens it will use. analyze.py creates and writes to ~/.openclaw/workspace/stock_data_output (file‑write behavior), so the skill will touch user home files despite declaring no config paths. The absence of declared credentials for cloud APIs is an inconsistency.
Persistence & Privilege
always is false and there is no install script that modifies other skills or system settings. The skill does create a per‑user workspace directory (~/.openclaw/workspace/stock_data_output) for outputs, which is typical for such tools. Autonomous invocation is allowed by default (not flagged), but the skill does not request elevated or persistent platform privileges.
What to consider before installing
Do not install or enable this skill until the author clarifies missing components and required credentials. Specifically: (1) confirm and provide the missing sub-scripts (scripts/a-stock/, scripts/glmv/, scripts/autoglm/) or update SKILL.md to match the included code; (2) specify exactly which environment variables or API tokens (Feishu, Tushare, data providers) are required and what scopes/permissions they need; (3) inspect any additional scripts for external endpoints and credential usage before providing secrets; (4) be aware the skill will write output files to ~/.openclaw/workspace/stock_data_output; run in a sandbox or test account first. The current package appears incomplete rather than overtly malicious, but the inconsistencies could allow unexpected behavior if missing components are later pulled from external sources.

Like a lobster shell, security has layers — review code before you run it.

latestvk970k9z4fdgr1kbx6gdn6pyssn846p29
91downloads
0stars
1versions
Updated 3w ago
v1.0.0
MIT-0

股票投资智投顾问

技能概述

整合五大股票投资技能的综合智投顾问系统,提供从数据采集、多维度分析到专业报告输出的全流程服务。

核心能力:

  • 实时数据采集:沪深北港美股实时行情、分时量能、资金流向
  • 多源信息搜索:新闻、财报、公告、分析师评级
  • 多模态图表分析:K线图/分时图智能识别与解读
  • 四角色协作分析:市场研究、宏观研判、情绪分析、投资决策
  • 专业报告输出:飞书云文档,咨询级格式

触发条件

  • "分析一下XX股票" / "XX股票怎么样"
  • "XX能不能买" / "XX该不该卖"
  • "生成XX的投资报告" / "XX的投资建议"
  • "比较XX和YY" / "XX板块分析"
  • 发送股票K线图/走势图截图
  • 查询持仓盈亏/投资组合分析

支持市场

市场代码格式示例
A股(沪深)6位数字600519、002446
A股(北交所)8开头830799
港股数字.HK0700.HK、09988.HK
美股字母代码AAPL、TSLA

工作流程

Phase 1: 标的确认与数据采集
├─ Step 1: 确认标的代码(web_search)
├─ Step 2: 实时行情采集
├─ Step 3: 分时量能分析
└─ Step 4: K线图/资金流图采集

Phase 2: 多源信息搜索
├─ Step 5: 新闻搜索
├─ Step 6: 财报/公告获取
├─ Step 7: 分析师评级搜索
└─ Step 8: 宏观与行业信息

Phase 3: 四角色协作分析
├─ Step 9: 市场研究分析师(板块+量化评分)
├─ Step 10: 经济顾问(宏观研判)
├─ Step 11: 投资者行为分析师(情绪+资金)
└─ Step 12: 投资组合经理(整合研判)

Phase 4: 报告生成与输出
├─ Step 13: 生成飞书云文档报告
└─ Step 14: 输出精炼总结

Phase 1: 标的确认与数据采集

Step 1: 确认标的代码

不论你是否认得这只股票,都必须先搜索一次确认代码。

web_search: "{用户说的名字} 股票代码 上市 港股 OR A股 OR 美股"

Step 2: 实时行情采集

使用 a-stock-analysis 的 analyze.py 脚本:

uv run {SKILL_DIR}/scripts/a-stock/analyze.py {股票代码}
uv run {SKILL_DIR}/scripts/a-stock/analyze.py {股票代码} --minute
uv run {SKILL_DIR}/scripts/a-stock/analyze.py {股票代码} --json

Step 3: 分时量能分析

时段说明信号
早盘30分主力早盘动作>30%为抢筹信号,>40%强势介入
尾盘30分尾盘异动>15%放量,>25%抢筹/出货

Step 4: K线图/资金流图采集

{SKILL_DIR}/scripts/glmv/venv/bin/python {SKILL_DIR}/scripts/glmv/fetch_all.py {股票代码}

输出文件: kline_em.png, kline_intraday.png, capital_flow.png, data.json

⚠️ 必须用 read 工具查看每张图片!


Phase 2: 多源信息搜索

Step 5-8: 信息搜索

使用 web_search 至少搜索 2-3 次:

web_search: "{股票名称}" "{股票代码}" 最新 2026年
web_search: "{股票名称}" 财报 业绩 营收
web_search: "{股票名称}" 分析师 评级 目标价 研报
web_search: "{行业名}" 板块 景气度 2026

新闻筛选原则:

  • ✅ 直接提到该股票名称或代码
  • ✅ 该公司公告/财报/业绩指引
  • ✅ 直接影响该标的的行业政策
  • ❌ 全市场无关快讯
  • ❌ 泛宏观新闻

Phase 3: 四角色协作分析

Step 9: 市场研究分析师

职责:板块研究 + 量化评分 + 龙头追踪

热度评级

评级标准操作建议
🔥🔥🔥🔥🔥近5日涨幅>15%顺势持有,注意高位风险
🔥🔥🔥🔥近5日涨幅10-15%逢低配置
🔥🔥🔥近5日涨幅5-10%关注催化持续性
🔥🔥涨幅0-5%等待机会
🔥平淡或小幅流出观望

量化综合评分(1-10分制)

维度权重评分依据
盈利增速30%业绩增速、ROE、毛利率
资金流向30%主力净流入、北向资金、融资余额
技术形态20%趋势强度、均线排列、支撑压力
催化剂20%政策/事件/业绩催化强度

Step 10: 经济顾问

职责:宏观研判 + 政策分析 + 海外因素

分析国内宏观环境(GDP、LPR、PMI、社融)和海外影响因素(美联储利率、中美关系、地缘政治)。


Step 11: 投资者行为分析师

职责:情绪分析 + 资金动向 + 行为策略

资金指标分析

指标解读
主力净流入主力态度
北向资金外资动向
融资余额杠杆资金情绪
换手率活跃度

行为策略建议

策略类型适用场景操作要点
顺势策略上升趋势确认回撤买入,不追高
反向策略高位震荡期分批止盈,控制贪婪
观望策略方向不明控制仓位,等待信号

Step 12: 投资组合经理

职责:整合研判 + 投资决策 + 风险管理

多空对比

🟢 做多逻辑🔴 做空逻辑
因素1(具体数据支撑)因素1(具体数据支撑)
因素2(具体数据支撑)因素2(具体数据支撑)

投资评级

维度评分
基本面⭐⭐⭐⭐
技术面⭐⭐⭐
市场情绪⭐⭐⭐
综合评级🟢买入/🟡观望/🔴回避

仓位管理

  • 单只标的仓位上限:8%(高波动股5-6%)
  • 止损线:亏损X%减仓/止损
  • 止盈策略:第一目标XX%,第二目标XX%

Phase 4: 报告生成与输出

Step 13: 生成飞书云文档报告

使用 feishu_create_doc 创建专业投资报告。

报告标题格式: [股票名称]([代码])投资分析报告_[日期]

报告结构:

  1. 执行摘要
  2. 基本信息
  3. 市场研究分析
  4. 宏观环境分析
  5. 技术面分析(含真实K线图)
  6. 资金与情绪分析
  7. 基本面分析
  8. 近期事件与新闻
  9. 投资建议
  10. 风险提示
  11. 免责声明

Step 14: 输出精炼总结

📊 [股票名称]([代码])快速总结

═══ 核心数据 ═══
**股价:** XX.XX元(今日±X.XX%)
**市值:** ~XX亿
**关键财务:** 营收XX亿 / 净利XX亿 / PE XX倍
**资金面:** 主力净流入XX亿

═══ 多空对比 ═══
| 🟢 做多逻辑 | 🔴 做空逻辑 |
|-----------|-----------|
| 因素1 | 因素1 |

═══ 结论与建议 ═══
**总评级:** 🟢买入 / 🟡观望 / 🔴回避
📄 完整报告:[飞书文档链接]

特殊场景处理

场景A:用户发送K线图截图

  1. 使用 image 工具识别图片
  2. 分析K线图:趋势、关键价位、技术形态
  3. 补充搜索该股票基本面信息

场景B:多只股票对比

  1. 对每只股票分别执行 Step 1-8
  2. 制作对比表格
  3. 给出选择建议

场景C:板块/行业分析

  1. 确认板块龙头股(3-5只)
  2. 分别分析各龙头
  3. 推荐优选标的

质量标准

  • 数据时效性:行情实时,新闻近3日,财报最新披露
  • 分析严谨性:所有结论有数据支撑,多空因素对等呈现
  • 报告规范性:飞书云文档格式,咨询级专业风格

注意事项

  1. 先确认代码再分析 - 永远不用记忆中的代码
  2. 必须亲自看图 - 多模态能力是核心价值
  3. 新闻要精准 - 不相关的全市场快讯不要塞进报告
  4. 数据不编造 - 没获取到就说暂无数据
  5. 风险提示充分 - 每份报告必须包含具体风险提示
  6. 免责声明完整 - 报告末尾必须包含免责声明

技能版本: v1.0
创建日期: 2026年04月03日
整合来源: 市场研究报告生成器 + AI投资顾问团队 + A股实时行情分析 + AutoGLM股票分析 + GLM-V股票分析师

Comments

Loading comments...