ftshare-lab/ftshare-skill

Overview

Making financial data a reliable context for AI. FTShare Skill enables agents to automatically select and call basic financial data and FTShare's signature factor interfaces based on natural language questions. This repository provides an installable parent skill: ftshare-market-data. After the agent reads the parent skill, it selects the corresponding data interface from the internal child skills, executes the request through run.py, and reads the JSON. All requests require environment variable FTSHARE_API_KEY. FTShare Skill is a financial data access method provided by FTShare for agent runtime. After loading ftshare-market-data into runtimes like Claude Code, Codex, OpenClaw, etc., users can directly ask data questions in natural language, and the agent handles interface selection, parameter organization, and result reading. Place the complete ftshare-market-data directory into the Skill directory of the Agent runtime. The directory must fully retain SKILL.md, run.py, and sub-skills/. This interface requires the stock code to include the exchange suffix, such as 600519.SH. Feature factors are research data, and the specific usability depends on the account package and does not constitute stock recommendations or future earnings judgments. There are three ways to connect to the same FTShare financial data service. SDK is required for stable calls in Python programs; MCP is used when standard MCP tools are needed; Skills are used when agents are required to automatically select data interfaces based on the issue. Currently, Skill covers data domains including spot, macroeconomics, large model corpora, stocks, US stocks, public funds, ETFs, Hong Kong stocks, futures, bonds, and indices.

Overview (machine translated)

This page was translated automatically. Technical details may differ from the original.

Recommended Tools

Try a different keyword or remove a filter.

Install