Developer APIBeta

Model Context Protocol

Connect your local AI environment, quantitative trading algorithms, and custom agents directly to Meshflow's proprietary data via the open-source Model Context Protocol (MCP).

Built and maintained by Meshflow Acquisition Corp.

Built for Developers

No need to parse messy HTML or manage headless browser scrapers. Get structured, pristine JSON data directly into your terminal or AI IDE.

Agent Native

Exposes functions specifically designed for Autonomous Agents to understand financial intent—like searching entities and fetching normalized structural data.

Integration Guides

Claude Desktop

Map our secure HTTP SSE endpoint to standard I/O (stdio) using the official MCP inspector tool as a bridge.

claude_desktop_config.json
{
  "mcpServers": {
    "meshflow": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/inspector",
        "https://www.meshflow.com/api/mcp"
      ]
    }
  }
}

On macOS, this configuration file is located at ~/Library/Application Support/Claude/claude_desktop_config.json.

Cursor IDE

Cursor has native support for MCP servers, allowing you to seamlessly integrate Meshflow SPAC data directly into your coding environment (e.g., when building quantitative scripts).

  1. Open Cursor Settings ⌘ + ,
  2. Navigate to Features > MCP
  3. Click + Add New MCP Server
  4. Set the type to SSE
  5. Name: meshflow
  6. URL: https://www.meshflow.com/api/mcp

OpenAI / ChatGPT (Web)

While ChatGPT's web interface does not natively support live SSE MCP connections today, we have exposed our data via the emerging Agentic SEO standard.

Simply inform the ChatGPT web interface to read our context file:
https://www.meshflow.com/llms.txt
The OpenAI web crawler will ingest our context instructions, data structures, and entity maps directly from this file to inform its answers.

Available Tools

search_spacs

Searches for special purpose acquisition companies by name or ticker to resolve canonical data.

Arguments:
  • query (string): The company name or ticker
  • limit (number, optional): Max results, defaults to 5

get_spac_details

Retrieves comprehensive structural information for a specific SPAC including its management team, board members, and targeting metadata.

Arguments:
  • identifier (string): The canonical ticker symbol or exact name