Equity Research Skill — Análise de Ações Agressiva
gmanish10/equity-research-skill
Análise de equity research opinionada para carteiras agressivas com risco explícito
Ver no GitHubMCP server Python que oferece análise fundamentada de ações, ETFs, opções e portfólios através de uma metodologia de 7 fases (negócio, setor, fundamentals, técnica, opções, sentimento, veredito). Integra dados do Yahoo Finance com web search para pesquisa de risco e oportunidade em carteiras agressivas de crescimento, sempre com callouts de risco quantificados.
Instalação
npx convez-ai gmanish10/equity-research-skill1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Análise fundamentada de ações com rating multi-horizonte e risco quantificado
- Avaliação de portfólios (correlação, concentração, beta, drawdown, fatores)
- Geração de ideias de investimento via screens de mercado com workup completo
- Análise de opções (IV rank, skew, fluxo, max pain, movimentos implícitos)
- Revisão de ETFs e fit estratégico na carteira existente
- Planos de rebalanceamento com gatilhos, prazos e callouts de risco por linha
Veredito

Ferramenta robusta e opinada para quem busca análise de equity research em Python com metodologia estruturada e risco explícito. Requer setup técnico (Yahoo Finance MCP, deps Python), disciplina para interpretação (não é aconselhamento personalizado) e perfil de risco agressivo. Disclaimer claro: dados com ~15 min de delay, recomendações educacionais apenas.
README
A tradução para o português deste README ainda está na fila.
Equity Research Plugin (Aggressive-Growth Edition)
Disclaimer — read this first. This plugin produces opinionated investment analysis for educational and informational purposes only. It is not personalized financial advice. Nothing produced by this plugin should be taken as a recommendation to buy, sell, or hold any security. Data is delayed ~15 minutes. Consult a qualified financial advisor before making investment decisions. Past performance does not predict future returns. Options involve substantial risk and can expire worthless. Leveraged ETFs are not long-term investments. You can lose money. Use at your own risk.
A Claude Code plugin for opinionated, aggressive-growth equity research — stocks, options, ETFs, sectors, and full portfolio reviews. Uses the Yahoo Finance MCP for quantitative data and web search for qualitative context.
What it does
- Deep stock research — a 7-phase workup (business, sector, fundamentals, technicals, options, sentiment, verdict) culminating in a multi-horizon rating with explicit, quantified risk callouts for short / mid / long-term horizons.
- Portfolio analysis — intake from Excel, CSV, screenshots, broker PDFs, or typed text. Normalized, confirmed back, then run through concentration, correlation, beta, drawdown, and factor analysis.
- ETF analysis — methodology, holdings, factor exposure, correlation with your existing book, role fit.
- Sector briefs — composition, aggregates, regime fit, best-in-class names, entry vehicles.
- Options flow — IV rank, skew, put/call ratios, max pain, magnet strikes, implied moves, unusual activity.
- Rebalance plans — concrete action matrices with triggers, deadlines, and risk callouts per line.
- Idea generation — screens live market data for new stocks, ETFs, or options overlays to fill a mandate or a gap in your book — candidates are sourced from a screen and run through a real workup, never pulled from memory.
The framing
This plugin defaults to an aggressive growth, higher-than-average risk tolerance lens. It's willing to take a stance. It considers higher-beta names, thematic plays, leveraged ETFs (with warnings), and options overlays.
But every aggressive recommendation carries a mandatory, quantified risk callout — a WHAT KILLS THIS TRADE block specifying exactly what would invalidate the thesis and how much it would cost. Aggressive != reckless.
Install on Claude Code
This repo includes Claude Code plugin marketplace metadata. Install it with two commands from inside Claude Code:
/plugin marketplace add gmanish10/equity-research-skill
/plugin install equity-research@comprehensive-equity-research
After install, restart Claude Code to pick up the new slash commands.
Prerequisites
-
Claude Code with plugins enabled
-
Yahoo Finance MCP server configured
-
Python 3.10+ on the machine running Claude Code
-
Python packages:
pandas numpy openpyxl rapidfuzz yfinance python-docxpip install pandas numpy openpyxl rapidfuzz yfinance python-docx
Install as a standalone skill
If you're not using a plugin-capable client, you can install just the skill (without the slash commands) by packaging or copying plugins/equity-research/skills/equity-research/ into your client's skill location. In clients that support .skill files, use the release asset if one is published.
Usage
Slash commands exposed by the plugin:
| Command | What it does |
|---|---|
/research-stock TICKER | Full 7-phase deep dive |
/analyze-portfolio [file] | End-to-end portfolio review with intake |
/analyze-etf TICKER | ETF one-pager |
/sector-brief SECTOR | Sector analysis + best-in-class picks |
/options-flow TICKER | Options setup read |
/rebalance-plan | Action matrix after a portfolio review |
/find-ideas MANDATE | Screen live data for new buy ideas (stocks, ETFs, overlays) |
Or just ask naturally — the skill's description triggers automatically on investment-related queries:
- "Do a deep research workup on NVDA. I want short/mid/long-term views."
- "Here's my portfolio (attaches file). Review it with an aggressive growth lens."
- "What's the options setup on TSLA for next month?"
- "Give me a sector brief on semiconductors."
- "Find me some new AI-exposure ideas that aren't already in my portfolio."
- "Based on the review, what should I actually do this week?"
Directory structure
equity-research-skill/
|-- .claude-plugin/
| \-- marketplace.json # Claude Code marketplace manifest
|-- plugins/
| \-- equity-research/ # Plugin root
| |-- .claude-plugin/
| | \-- plugin.json # Claude plugin manifest
| |-- commands/ # Slash commands
| | |-- research-stock.md
| | |-- analyze-portfolio.md
| | |-- analyze-etf.md
| | |-- sector-brief.md
| | |-- options-flow.md
| | |-- rebalance-plan.md
| | \-- find-ideas.md
| \-- skills/
| \-- equity-research/ # The skill
| |-- SKILL.md # Skill entry point
| |-- references/ # Deep-dive methodology docs
| |-- scripts/ # Python - technicals, ratios, portfolio intake, etc.
| \-- evals/ # Eval test cases
|-- README.md
|-- LICENSE
|-- DISCLAIMER.md
\-- .gitignore
Credits
Built on top of Anthropic's bundled stock-analyst skill. The Yahoo Finance MCP orchestration patterns are adapted from that foundation. Specialization (multi-horizon ratings, aggressive-growth framing, ETF workflow, multi-format portfolio intake, risk-callout mandate) is original to this plugin.
Contributing
Open an issue or PR. Useful directions:
- Additional broker-statement parsers in
plugins/equity-research/skills/equity-research/scripts/parse_portfolio.py - More sophisticated DCF / fair value modeling
- Backtesting harness for rating accuracy
- Earnings-call transcript summarization
- Alternative data integrations (e.g., short-seller report tracking)
License
MIT. See LICENSE.
One more time - this is not financial advice
The plugin is opinionated by design. That does not mean its opinions are correct. Do your own research. Consult a professional. Size positions appropriately. Know what you can afford to lose. Markets are adversarial and humbling.