wb-finance-skill

Analyze stocks, funds, macro data, and portfolios using routed financial data retrieval and scenario methodology references.

68|25|Updated Jul 24, 2026
One-click install
npx skills add https://github.com/rojim666/SztuCode --skill wb-finance-skill-rojim666
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: wb-finance-skill
Source: https://github.com/rojim666/SztuCode/tree/main/py-runtime/src/sztu_code/core/prompts/workbuddy/skills/wb-finance-skill
Command: npx skills add https://github.com/rojim666/SztuCode --skill wb-finance-skill-rojim666

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, requests, openpyxl, pyharmonics, czsc, smartmoneyconcepts, and includes scripts (resource) and references (resource) components.

What problem does it solve? Financial questions are easy to answer badly: fabricated numbers, stale quotes, mismatched fiscal periods, and unsourced claims. This Skill enforces a disciplined workflow for any finance-related request—mandating tool-based data retrieval instead of memory, source and timestamp attribution for every key figure, cross-source verification, and fixed disclaimers whenever actionable advice is given. ## Core Features & Use Cases - Scenario-routed methodology: Maps user questions (stock research, valuation, earnings review, trade plans, sector rotation, macro transmission, options, quant factors, IB modeling) to 40+ dedicated reference playbooks with thresholds and pitfalls. - Data retrieval discipline: Delegates market, financial, and macro queries to an agentic search tool or Tongdaxin MCP with strict rules on query phrasing, time-period resolution, and source quality grading. - Executable analysis engines: Ships price-action signal scripts (candlestick, harmonic, Elliott wave, Chan theory, Ichimoku, SMC, VCP) and quant strategy engines (pair trading, seasonality, volatility, multi-factor) plus IB utilities (DCF validation, number consistency checks). - Use Case: A user asks "Should I add to my position in a stock down 40%?" The Skill loads the position-sizing and stop-discipline references, pulls current quotes and fundamentals via search tools, de-anchors from cost basis, and delivers a conditional action framework with the mandatory disclaimer. ## Quick Start Ask any finance question such as "analyze whether Kweichow Moutai is worth researching right now" and the Skill will route it, retrieve live data, and produce a sourced analysis.

Frequently Asked Questions about wb-finance-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a stock with live market data instead of model memory?▼

Delegate the lookup to the agentic search tool with a single-sentence intent naming the ticker and topic, then apply the matching methodology reference such as stock-deep-research or valuation-pricing. Every key figure must carry a source and timestamp, and memory-based numbers are prohibited.

What technical analysis tools are included for K-line and pattern recognition?▼

The scripts directory provides seven signal engines covering candlestick patterns, harmonic patterns, Elliott wave, Chan theory, Ichimoku, smart money concepts, and basic indicators like MACD and RSI. Run them via the run_signal.py CLI with a data source and ticker code.

Does the skill support Tongdaxin MCP for A-share data?▼

Yes, when the Tongdaxin MCP tools are available in the environment, the skill routes structured queries for quotes, K-lines, financial statements, shareholder data, and screeners through them. The tdx-mcp-quick-reference file documents verified call patterns, fixedTag routing, and known limitations.

How are cross-market fiscal year comparisons handled?▼

Fiscal year labels are never compared directly because companies like Tencent, Alibaba, Apple, and Microsoft use different fiscal calendars. Each fiscal period is mapped to its actual calendar quarters, and TTM rolling four-quarter sums are used for annual comparisons.

Why must investment advice outputs include a fixed disclaimer?▼

Any output containing buy/sell calls, price levels, or position adjustments must append a verbatim disclaimer stating the content is not investment advice. This is a hard rule that cannot be reworded, shortened, or omitted, ensuring consistent risk communication.

What are the limitations of the quant strategy scripts?▼

The six quant engines cover pair trading, seasonality, volatility, multi-factor, fundamental screening, and minute-level data, but event-driven, machine learning, and hedging strategies have no executable scripts and must be implemented from the methodology references. All scripts require pandas, numpy, and requests.