Co-Invest MCP Server ★ New (July 3)¶

Overview¶

Co-Invest MCP (co-invest-mcp) enables AI agents to participate in investment research and management workflows. Built on Liquid Trade's investment infrastructure, this MCP server lets agents analyze markets, research assets, and execute investment strategies — all within the agent conversation. Official MCP from Liquid Trade.

Key advantage: Investment management becomes conversational. Instead of switching between Bloomberg, brokerage apps, and spreadsheets, operators can research and manage investments directly through their AI agent.

Key Features¶

Market research: Query real-time and historical market data across asset classes
Portfolio analysis: Review holdings, performance metrics, and risk exposure
Trade execution: Place and manage investment orders (with confirmation gates)
Strategy backtesting: Test investment strategies against historical data
Alert monitoring: Set and monitor price alerts, news triggers, and portfolio thresholds
Web-based platform: Accessible at liquid.trade/coinvest

Installation¶

# Add to Hermes (remote endpoint)
hermes mcp add coinvest --url https://api.liquid.trade/coinvest/mcp

# Authenticate
hermes mcp auth coinvest

Manual Configuration¶

{
  "mcpServers": {
    "coinvest": {
      "type": "sse",
      "url": "https://api.liquid.trade/coinvest/mcp",
      "headers": {
        "Authorization": "Bearer ${LIQUID_TRADE_API_KEY}"
      }
    }
  }
}

Prerequisites¶

Liquid Trade Account: Sign up at liquid.trade/coinvest
API Key: Generate from account settings
Linked Brokerage (optional): Connect brokerage for trade execution

Business Use Cases¶

Research Assistant: "Compare the PE ratios and revenue growth of these 5 SaaS companies" — agent pulls live data
Portfolio Review: "Show me my portfolio allocation and flag any positions over 20% concentration"
Alert Monitoring: Agent monitors positions and alerts you when a stock hits your target price
Due Diligence: "Research this company — management team, financials, competitive position, SEC filings"
Strategy Execution: Define investment rules and let the agent monitor for opportunities matching your criteria

Business Relevance¶

Investment management has been fragmented across terminals, apps, and spreadsheets. Co-Invest MCP brings it into the AI agent workflow — operators managing personal or firm investments can research, monitor, and execute without context-switching. The "invest with Claude" paradigm signals where agentic finance is heading.

Limitations¶

Trade execution requires confirmation gates (by design — safety feature)
Market data coverage varies by asset class and exchange
Investment advice is AI-generated — not a substitute for professional financial advice
US-focused initially with international expansion planned
Beta product — API and tool surface may change