← Back to Skills Marketplace
133
Downloads
1
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install stock-analysis-china
Description
A股持仓深度技术分析技能。当用户发送持仓截图图片、或提及持仓/股票/投资建议相关意图时，本技能自动激活。 激活条件（满足任一即触发）： - 用户发送了图片（持仓截图、行情截图等） - 用户说"分析持仓"、"我的持仓怎么样"、"给投资建议"、"持仓诊断"、"我的股票"、"目前持仓" - 用户说"更新持仓"、"录入持...
Usage Guidance
This skill appears to do what it says: read screenshots (or a local portfolio JSON), call AKShare for market data, compute indicators, and generate a report. Things to consider before installing: (1) Privacy — the skill reads your screenshot folder (~/.qclaw/media/inbound/) and will store portfolio data under the skill's data directory; review or back up existing positions_portfolio.json before use. (2) Dependencies — the Python scripts require akshare and pandas; SKILL.md also suggests installing Tesseract (for OCR fallback) and optionally Node.js + sharp for image handling even though the shipped Python code does not use Node. You only need to install Node/sharp if you plan to use that alternate image-processing path. (3) Run review — inspect scripts yourself (they are small and local) or run them in a sandbox/virtual environment if you are concerned. (4) No credentials are requested and there are no hidden network endpoints beyond AKShare (a public market-data library), so the main risks are privacy of screenshots and granting system installs (Tesseract/Node) — accept those only if you trust the environment.
Capability Analysis
Type: OpenClaw Skill
Name: stock-analysis-china
Version: 0.1.1
The skill bundle facilitates A-share stock analysis but requires high-risk capabilities, including the execution of shell commands (Python, Tesseract, npm, winget) and broad file system access to the user's home directory (~/.qclaw). SKILL.md and the analysis template contain imperative instructions that force the AI agent to run these commands and perform environment modifications, such as creating directory junctions and installing global packages. While the Python scripts (stock_analysis_core.py and portfolio_update.py) appear focused on legitimate financial data processing via AKShare, the reliance on shell-based automation and the potential for command injection via file paths or OCR output warrants a suspicious classification.
Capability Assessment
Purpose & Capability
The skill's name/description (A股持仓技术分析) matches the actions in SKILL.md and the Python scripts: reading screenshots or a portfolio JSON, updating local positions, fetching market data via AKShare, and computing technical indicators. One mismatch: SKILL.md lists Node.js and the 'sharp' package for image processing, but included code files are pure Python and do not call Node/Sharp; this dependency appears unnecessary for the provided scripts (it may be a suggested optional tool for alternate image handling).
Instruction Scope
SKILL.md instructs the agent to (a) read recent images from ~/.qclaw/media/inbound (or ~/.openclaw/media/inbound), (b) prefer AI multimodal recognition and fall back to Tesseract OCR, (c) update SKILL_ROOT/data/positions_portfolio.json, and (d) run the Python analysis script. These actions are consistent with the stated purpose. Note: reading user screenshots and updating a local JSON are privacy-sensitive but expected for this skill.
Install Mechanism
No install spec is provided (instruction-only), and the code is purely local Python scripts — low install risk. SKILL.md suggests installing akshare/pandas, Tesseract, Node.js and sharp; the Python dependencies (akshare/pandas) are required by the analysis script, while Node.js/sharp are only suggested for image processing but are not used in the Python code. No downloads from untrusted URLs or archive extraction are present.
Credentials
The skill requests no environment variables or external credentials — appropriate. It does recommend installing system tools (Tesseract) and optionally Node.js/global sharp; those require system-level installation/privileges but are not required by the included Python scripts. The script accesses local paths (user media inbound and SKILL_ROOT/data) which is expected for screenshot-based functionality; there is no evidence it accesses unrelated credentials or external secrets.
Persistence & Privilege
The skill does not request always:true and uses normal autonomous invocation defaults. It writes and reads its own data file under the skill directory (data/positions_portfolio.json) and reads the user's screenshot inbound directory — this is within expected scope. It does not modify system-wide agent config or other skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install stock-analysis-china - After installation, invoke the skill by name or use
/stock-analysis-china - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.1
A股持仓深度技术分析，具备如下几个特点：
1. 使用简便：只需将微信/飞书持仓截图发送至Qclaw/openclaw并通过脚本生成专属深度技术分析报告
2. 泛用性强：同时支持Qclaw和openclaw，其中Qclaw主要支持微信，openclaw支持一切图片来源
3. 智能运行：若环境缺失，第一次运行自动配置部分环境（OCR等）
使用注意事项：第一次执行会自动安装部分环境配置，可能等待时间较长
Initial release of stock-analysis-china.
- Provides A股持仓深度技术分析 via AI, activating on screenshots or investment-related intents.
- Supports AI-based image recognition of WeChat/Feishu portfolio screenshots, with fallback to Tesseract OCR.
- Manages portfolio data in positions_portfolio.json, auto-updating from images or user input.
- Runs technical analysis with AKShare (RSI, MACD, KDJ, Bollinger Bands, moving averages) and generates operation suggestions.
- Outputs structured analysis reports directly in chat, including market index overview and individual stock signals.
- Includes detailed environment setup and troubleshooting guidance for common installation and runtime issues.
v0.1.0
stock-analysis-china 0.1.0 初始版本发布
- 支持微信A股持仓截图自动识别（多模态/OCR双重识别），自动提取股票名称、代码、数量、成本价等信息
- 持仓数据自动管理，支持自动更新和手动录入
- 集成AKShare，实时获取大盘及个股行情，支持RSI/MACD/KDJ/布林带/均线等主流技术指标分析
- 按序执行：识别图片→读取/更新持仓→技术分析→结构化报告输出（含操作建议）
- 支持多种激活方式：上传图片、关键词触发、持仓/投资建议等场景一键分析
- 提供详细依赖、路径兼容说明及常见故障排查指引
Metadata
Frequently Asked Questions
What is Stock Analysis China?
A股持仓深度技术分析技能。当用户发送持仓截图图片、或提及持仓/股票/投资建议相关意图时，本技能自动激活。 激活条件（满足任一即触发）： - 用户发送了图片（持仓截图、行情截图等） - 用户说"分析持仓"、"我的持仓怎么样"、"给投资建议"、"持仓诊断"、"我的股票"、"目前持仓" - 用户说"更新持仓"、"录入持... It is an AI Agent Skill for Claude Code / OpenClaw, with 133 downloads so far.
How do I install Stock Analysis China?
Run "/install stock-analysis-china" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Stock Analysis China free?
Yes, Stock Analysis China is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Stock Analysis China support?
Stock Analysis China is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Stock Analysis China?
It is built and maintained by lzwaang (@lzwaang); the current version is v0.1.1.
More Skills