A multi-agent financial market analysis and trading intelligence suite built natively on Claude Code's ecosystem — MCP servers, skills, hooks, subagents, and rules.
-
Clone and open in Claude Code:
cd firemonkey claude -
Set up API keys:
cp .env.example .env # Edit .env with your API keys -
Use the agents:
/market-analyst AAPL # Technical + fundamental analysis /news-sentiment TSLA # News and sentiment report /trade-strategist NVDA long swing # Generate trade setup /risk-manager # Validate a trade proposal /portfolio-manager # View holdings and P&L /trade-executor # Execute an approved trade /morning-brief # Daily market overview
| Layer | Location | Purpose |
|---|---|---|
| Skills | .claude/skills/ |
7 specialist agent definitions |
| Rules | .claude/rules/ |
Risk limits, conventions, security |
| Hooks | .claude/settings.json |
Automation, validation, audit |
| MCP Servers | .mcp.json |
Financial data sources |
| Strategies | strategies/ |
Trading strategy playbooks |
| Templates | templates/ |
Report format standards |
| Data | data/ |
Portfolio state, history, watchlists |
| Server | Data |
|---|---|
| Alpha Vantage | Real-time quotes, technicals, forex, crypto |
| Finnhub | 50+ indicators, news, ESG, insider trades |
| Financial Datasets | Financial statements, market news |
| Alpaca | Paper trading, order management |
| YFinance | Company overviews, analyst data |
- Alpha Vantage — Free
- Finnhub — Free
- Alpaca — Free paper trading
- Financial Datasets — Free tier
- Paper trading by default — Alpaca is configured to paper endpoint
- Human confirmation required for all trade executions
- Risk limits enforced by hooks and the Risk Manager agent
- Audit logging of all trade activity
- No secrets in code — all credentials via environment variables
MIT