fdic-bank-find

The FDIC BankFind MCP Server exposes a set of endpoints that mirror the FDIC BankFind API, wrapped to work with MCP-compatible tooling. It lets you query structured U.S. banking data through a consistent MCP interface, enabling you to drive AI agents, dashboards, and analytics pipelines with ready-made endpoints such as get_demographics, get_institutions, get_locations, get_summary, get_sod, get_history, and get_failures. Each tool accepts common parameters like api_key, filters, fields, limit, offset, sort_by, sort_order, and file_format to customize results and formats. The MCP Inspector can be used to test and inspect requests/responses against your running server, providing a visual interface for debugging and prototyping. Use the docker-based deployment to start the server, then configure your MCP host (e.g., in VS Code or Claude Desktop) to point to the running container. The server is designed to be lightweight and suitable for AI workflows, with endpoints that expose institutional demographics, historical aggregates, and FDIC BankFind data.

Typical usage flow:

• Start the server via Docker using the published image.
• Point your MCP host to the server using the provided mcp.json blocks for your environment.
• Call endpoints like get_institutions (for institution demographics) or get_demographics (for demographic summaries) with optional filtering, field selection, and pagination.
• Combine results with your agent or analytics tools, and iterate as needed using the MCP Inspector for debugging.

Prerequisites:

• Docker installed and running on your machine (recommended).
• Basic familiarity with MCP configuration in your hosting environment (VS Code MCP settings or Claude Desktop config).

Installation steps:

1. Pull and run the official public Docker image (recommended):

# Run the MCP server from GHCR
docker run -i --rm ghcr.io/clafollett/fdic-bank-find-mcp-server:main

2. If you prefer to build/run locally from source (advanced):

• Clone the repository and build the Rust binary (optional, for custom development):

git clone https://github.com/YOUR-ORG/fdic-bank-find-mcp-server.git
cd fdic-bank-find-mcp-server
cargo build --release

• Run the built binary with stdio mode or containerize as needed.

3. Configure your MCP host to invoke the server:

• In VS Code, add the following to your User Settings (JSON) or Workspace settings (without the outer mcp key for workspace):

{
  "mcp": {
    "servers": {
      "fdic": {
        "command": "docker",
        "args": [
          "run",
          "-i",
          "--rm",
          "ghcr.io/clafollett/fdic-bank-find-mcp-server:main"
        ]
      }
    }
  }
}

• For Claude Desktop or other MCP clients, use a similar block pointing to the same docker image:

{
  "mcpServers": {
    "fdic-bank-find": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/clafollett/fdic-bank-find-mcp-server:main"
      ]
    }
  }
}

4. Optional: run the MCP Inspector to test and visualize requests:

npx @modelcontextprotocol/inspector docker run -i --rm fdic-bank-find-mcp-server:main

Notes and tips:

• This server exposes all FDIC BankFind endpoints except /financials due to complexity and tooling limitations in Rust code generation. If you need that portion, you may need to experiment with workarounds or future updates.
• The recommended deployment path is the official Docker image; building from source is mainly for development or customization.
• When running in containers, ensure networking allows your MCP host to reach the container if you’re binding to a local port or using containerized IPC.
• The MCP Inspector is a helpful debugging aid; use it to validate request/response payloads and to prototype tool parameters before integrating into production workflows.
• Endpoints support common BankFind parameters (filters, fields, limit, offset, sort_by, sort_order, file_format, file_download, file_name) to tailor results.