MobiusQuant/Gendangzou-skill

What it solves

This project provides AI Agents with specialized capabilities to analyze A-share market sectors. It solves the problem of bridging the gap between raw financial data (policies, media reports, and capital flows) and actionable sector-level insights, ensuring that every conclusion is backed by a traceable evidence chain rather than generic AI hallucinations.

How it works

gendangzou-skill acts as a standardized capability package (Skill) that can be integrated into various AI Agent platforms (such as Claude Code, Codex, and OpenClaw). It connects the Agent to the Gendangzou API via a unified SDK, allowing the Agent to query:

• Real-time Discovery: Immediate tracking of new policies and authoritative media materials.
• Stable Research: Daily snapshots of sector heat, market attention, and capital confirmation.
• Relationship Auditing: Traceable mappings between companies/ETFs and specific sectors.
• Application Orchestration: Pre-built apps for daily/weekly reports and CCTV news interpretation.

Who it’s for

• AI Agent Developers: Those building financial assistants or research workflows who need a plug-and-play data layer for A-shares.
• Financial Analysts: Users who want to use AI Agents to monitor policy shifts and sector dynamics with verifiable sources.
• Quantitative Researchers: People needing structured, API-driven access to sector-level market signals.

Highlights

• Traceable Evidence: Every judgment includes sources, dates, and stable IDs to prevent hallucinations.
• Platform Agnostic: Designed to work across multiple Agent platforms using a shared SKILL.md and API SDK.
• Multi-layered Data: Distinguishes between high-frequency real-time materials and reproducible daily snapshots.
• Built-in Research Apps: Includes specialized tools for sector radar, mapping audits, and interpreting official news broadcasts.