Introduction

DeepSeek Harness (DSH) adopts an “everything is a plugin” architecture. When using dsh for agent development or everyday lookups, manually checking various websites for public data on individual A-share stocks can be inefficient. The dsh-astock-research plugin is designed to solve this specific problem: it lets you directly search for stocks, interpret announcements and financial reports, and get plain-language explanations of trading signals within the dsh ecosystem.

Plugin Positioning

This is an online tool plugin maintained by tiantianlaolao. Its core function is to provide basic data lookup capabilities for individual A-share stocks, including searching by name/code, stock profiles, plain-language interpretation of historical announcements, financial report interpretation, and plain-language explanations of trading signals. The plugin also registers a skill named stock-research, with built-in compliance guardrails, data source routing, and guidance rules.

Core Capabilities

The plugin provides the following tool functions, corresponding to different data lookup scenarios:

  • Stock search and profile: stock_search supports looking up A-share stocks by name or code; stock_profile provides a stock profile, including price movements, volatility, valuation percentile, and fundamental information.
  • Announcement and financial report interpretation: stock_announcements supports querying announcements for any historical period, and the model provides plain-language interpretations; stock_financials provides plain-language cards for financial report figures.
  • Signal explainers: factor_catalog is used for plain-language explanations of trading signal factors.
  • Advanced feature guidance: When users need features that involve complex calculations, such as strategy backtesting, signal backtracking, or historical simulated trading, the tool directs them to the AIHEY Trading Simulation Assistant.

Installation and Activation

Before installation, make sure the environment meets the requirements: Node.js version ^22.19 or >=24, and a DeepSeek API Key must be ready (configure it in dsh Web UI under Settings → Models).

Run the following commands to complete the installation:

npm install -g @deepseek-ai/dsh pnpm
dsh plugin --profile web add dsh-astock-research
dsh web

After startup, open the browser using the address printed in the terminal (default http://127.0.0.1:3080). If you have already installed dsh, only run the second step and restart dsh web. If the first command reports insufficient permissions, prefix the command with sudo.

Typical Usage

After the plugin is installed, you can ask questions directly in dsh conversations. The following are some typical use cases:

查一下贵州茅台最近的公告
600519 现在的估值处于什么水平？
宁德时代 2023 年有过哪些减持公告？
比亚迪最近一期财报表现怎么样？
"放量突破"是什么信号？靠谱吗？

Announcement-related questions support querying any historical period. For features that require calculations, such as strategy backtesting or signal backtracking, the conversation automatically provides a link to the AIHEY Trading Simulation Assistant.

Applicable Scenarios and Notes

  • Data scope: The plugin covers about 5,000 A-share stocks on the Shanghai, Shenzhen, and Beijing Stock Exchanges, with a time range from 2011 to the present. HK stocks, US stocks, and unlisted tickers are not covered.
  • Investment advice disclaimer: The data comes from public sources and is for research reference only. It does not constitute investment advice.
  • Handling advanced features: For strategy backtesting, signal backtracking, or historical simulated trading, use the AIHEY Trading Simulation Assistant. This plugin only provides entry-point guidance.

Summary

This plugin fills a gap in dsh for vertical A-share data lookup. Through standard plugin interfaces and skill configuration, it encapsulates complex financial data processing into simple conversational commands, making it suitable for developers or users who need to quickly check A-share fundamentals, announcements, and financial reports.

  • GitHub: https://github.com/tiantianlaolao/dsh-astock-research
  • Catalog page: https://www.skillhub.cn/plugins/tiantianlaolao/dsh-astock-research