Invest Committee — Data skill for Claude Code
Institutional-grade six-stage stock research pipeline for retail investors, as a Claude Code Agent Skill.
How to install Invest Committee
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open HenryLinyy/invest-committee and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Invest Committee does
Institutional-grade six-stage stock research pipeline for retail investors, as a Claude Code Agent Skill. Taiwan/US/HK/China-A markets. Research only, never places orders. 散戶自用的機構級個股研究流程,僅供研究參考不下單。
Alternatives in Data
- Data Contract — This document defines which files belong to the system (auto-updatable) and which belong to the user (never to 32.5k ★
- A Stock Data — A股全栈数据工具包 · 十一层架构 · 54端点 · 19数据源 · 零鉴权 Full-stack China A-share data toolkit for AI agents — 11 layers, 54 end 9.4k ★
- Ars Full — ARS full pipeline — research → write → review → revise → finalize 4.5k ★
README
invest-committee (Investment Committee)
**English** | [繁體中文](README.zh-TW.md)
Turn “analyze TSMC” into a source-cited investment-committee memo: fundamentals, institutional consensus, technical levels, risk flags, and a position-sizing action plan.
An [Agent Skill for Claude Code](https://claude.com/claude-code) that helps retail investors research **specific stocks** with more discipline. It covers **Taiwan / US / Hong Kong / China A-share** markets and routes data sources by ticker. It supports research only — it never places orders.
**Language note:** reports are currently generated in **Traditional Chinese (zh-TW)**. The workflow and methodology are market-agnostic.
Three things you can rely on
- Every figure needs a source. Prices, financials, and flow data are cited; unavailable values are labeled as missing.
- Data gaps are visible. Stale data, source fallbacks, and insufficient candle history are disclosed instead of presented as a complete conclusion.
- Research never becomes execution. The skill does not call trading APIs; its output is research support, not investment advice.
Try this first (one stage)
In Claude Code, say:
Analyze NVDA technicals
This runs only the relevant technical-analysis stage, so it is a faster and lower-token way to see whether the workflow fits you. When you are ready, try a full deep dive with “Analyze NVDA in depth.”
What are you trying to do?
| Your goal | Say this in Claude Code |
|---|---|
| Research one stock from fundamentals through risk | “幫我分析台積電” / “Analyze NVDA in depth” |
| Check only technicals or risk | “看 2330 技術面” / “掃 NVDA 風險” |
| Check the health of a watchlist | “健檢自選股” |
| Compare two candidate stocks | “比較台積電和聯電” / “Compare TSMC vs UMC” |
| Build a Taiwan day-trading playbook | “力成當沖” / “6239 當沖劇本” |
A full analysis delivers a source-cited research report and an HTML
Related Skills
AlphaAnalyst Open Source Autonomous Equity Research Agent
Enter a US stock ticker, get an analyst grade memo with DCF valuation, peer comparables, news sentiment, and e
AI Sdr OS
Production-grade agentic AI SDR (sales development rep) system: a 7-stage pipeline (Find, Score, Research, Per
Task Resume
Resume an in-progress or crashed pipeline run for a Jira ticket or REQ-ID, using persisted state and real repo
Claude Deep Research Skill
Enterprise-grade deep research skill for Claude Code with 8-phase pipeline, source credibility scoring, and au
Underwriter OS
Underwriter OS: a one-paste Claude setup that turns Claude into a research and underwriting teammate for insti
Xvary Stock Research
Claude Code skill for public SEC EDGAR + market data: /analyze, /score, /compare. MIT.
Related Agents
Execution Trader
The only Bot on the desk that places, modifies or cancels Hyperliquid orders. Executes one approved ticket at
Fg Builder
Finance Guru Document & Artifact Builder (Alexandra Kim). Creates institutional-grade financial documents, rep
QA Coach
Prepares founders for tough VC Q&A sessions by simulating the hardest questions investors will ask, coaching o