> ## Documentation Index
> Fetch the complete documentation index at: https://sec-edgar-mcp.amorelli.tech/llms.txt
> Use this file to discover all available pages before exploring further.

# MCP Clients

> Compatible MCP clients and configuration guides

# MCP Clients

SEC Edgar MCP works with any MCP-compatible client. Here's a comprehensive list of available clients and how to configure them.

## Desktop Applications

### Claude Desktop

<Card title="Claude Desktop" icon="robot" href="https://claude.ai/download">
  **Platform:** Windows, macOS, Linux\
  **Type:** Official Anthropic desktop app\
  **Best for:** Individual users, AI-powered research

  Claude Desktop is the most popular MCP client, providing direct integration with Claude AI for interactive financial analysis.

  **Configuration file location:**

  * **macOS/Linux:** `~/.config/claude-desktop/config.json`
  * **Windows:** `%APPDATA%\Claude\claude-desktop\config.json`

  ```json theme={null}
  {
    "mcpServers": {
      "sec-edgar-mcp": {
        "command": "docker",
        "args": [
          "run", "-i", "--rm",
          "-e", "SEC_EDGAR_USER_AGENT=Your Name (email@example.com)",
          "stefanoamorelli/sec-edgar-mcp:latest"
        ]
      }
    }
  }
  ```
</Card>

### Continue

<Card title="Continue" icon="code" href="https://continue.dev">
  **Platform:** VS Code, JetBrains IDEs\
  **Type:** IDE extension\
  **Best for:** Developers, code-integrated analysis

  Continue brings AI assistance directly to your IDE with MCP support for contextual coding help.

  **Configuration:** Add to `.continue/config.json`:

  ```json theme={null}
  {
    "models": [...],
    "mcpServers": {
      "sec-edgar-mcp": {
        "command": "uvx",
        "args": [
          "--from", "git+https://github.com/stefanoamorelli/sec-edgar-mcp.git",
          "sec-edgar-mcp"
        ],
        "env": {
          "SEC_EDGAR_USER_AGENT": "Your Name (email@example.com)"
        }
      }
    }
  }
  ```
</Card>

### Cline (formerly Claude Dev)

<Card title="Cline" icon="terminal" href="https://github.com/cline/cline">
  **Platform:** VS Code\
  **Type:** VS Code extension\
  **Best for:** Autonomous coding tasks, file management

  Cline is an autonomous AI assistant that can create, edit, and execute code with human oversight.

  **Configuration:** Through VS Code settings or `settings.json`:

  ```json theme={null}
  {
    "cline.mcpServers": {
      "sec-edgar-mcp": {
        "command": "python",
        "args": ["-m", "sec_edgar_mcp.server"],
        "env": {
          "SEC_EDGAR_USER_AGENT": "Your Name (email@example.com)"
        }
      }
    }
  }
  ```
</Card>

## Command Line Tools

### MCP CLI

<Card title="MCP CLI" icon="terminal" href="https://github.com/modelcontextprotocol/mcp-cli">
  **Platform:** All platforms\
  **Type:** Command-line interface\
  **Best for:** Testing, debugging, automation

  Official command-line tool for testing and debugging MCP servers.

  ```bash theme={null}
  # Install MCP CLI
  npm install -g @modelcontextprotocol/cli

  # Connect to SEC Edgar MCP
  mcp connect "docker run -i --rm -e SEC_EDGAR_USER_AGENT='Test User (test@example.com)' stefanoamorelli/sec-edgar-mcp:latest"

  # List available tools
  mcp tools list

  # Call a tool
  mcp call lookup_company_cik '{"company_name": "Apple"}'
  ```
</Card>

### MCP Client SDK

<Card title="MCP Client SDK" icon="code" href="https://modelcontextprotocol.io/docs/concepts/clients">
  **Platform:** Python, TypeScript/JavaScript\
  **Type:** SDK for building custom clients\
  **Best for:** Custom integrations, automation

  Build your own MCP client using official SDKs.

  **Python example:**

  ```python theme={null}
  from mcp import Client
  import asyncio

  async def main():
      async with Client() as client:
          # Connect to SEC Edgar MCP
          await client.connect(
              command="python",
              args=["-m", "sec_edgar_mcp.server"],
              env={"SEC_EDGAR_USER_AGENT": "Bot (bot@example.com)"}
          )
          
          # List available tools
          tools = await client.list_tools()
          
          # Call a tool
          result = await client.call_tool(
              "lookup_company_cik",
              {"company_name": "Tesla"}
          )
          print(result)

  asyncio.run(main())
  ```
</Card>

## Web Applications

### LibreChat

<Card title="LibreChat" icon="message" href="https://github.com/danny-avila/LibreChat">
  **Platform:** Web (self-hosted)\
  **Type:** Multi-model chat interface\
  **Best for:** Teams, self-hosted deployments

  Open-source AI chat platform with MCP support for multiple models.

  **Configuration:** Add to `librechat.yaml`:

  ```yaml theme={null}
  mcp:
    servers:
      sec-edgar:
        command: docker
        args:
          - run
          - -i
          - --rm
          - -e
          - "SEC_EDGAR_USER_AGENT=LibreChat (admin@example.com)"
          - stefanoamorelli/sec-edgar-mcp:latest
  ```
</Card>

### Open WebUI

<Card title="Open WebUI" icon="browser" href="https://github.com/open-webui/open-webui">
  **Platform:** Web (self-hosted)\
  **Type:** Web interface for LLMs\
  **Best for:** Self-hosted AI assistants

  Feature-rich web UI for various LLMs with extensibility through tools and functions.

  **Note:** MCP support is in development. Check their documentation for updates.
</Card>

## Mobile & Emerging Clients

### Mobile MCP Clients

<Note>
  Mobile MCP clients are emerging. Check the [MCP ecosystem](https://modelcontextprotocol.io/ecosystem) for the latest mobile client developments.
</Note>

### Custom Integrations

<Tip>
  Any application can become an MCP client by implementing the [MCP protocol specification](https://modelcontextprotocol.io/specification). The protocol uses JSON-RPC 2.0 over stdio, SSE, or WebSocket transports.
</Tip>

## Configuration Best Practices

<AccordionGroup>
  <Accordion title="Environment Variables">
    Always set the `SEC_EDGAR_USER_AGENT` with your actual name and email:

    ```bash theme={null}
    SEC_EDGAR_USER_AGENT="John Doe (john.doe@company.com)"
    ```

    The SEC requires proper identification for API access.
  </Accordion>

  <Accordion title="Docker vs Direct Installation">
    **Docker (Recommended):**

    * Isolated environment
    * No Python version conflicts
    * Consistent across platforms

    **Direct Installation (pip/uvx):**

    * Lower resource usage
    * Faster startup time
    * Easier debugging
  </Accordion>

  <Accordion title="Connection Testing">
    After configuration, test your connection:

    1. Restart your MCP client
    2. Check if SEC Edgar tools are discovered
    3. Try a simple query like company lookup
    4. Verify tool responses include SEC URLs
  </Accordion>

  <Accordion title="Troubleshooting">
    Common issues and solutions:

    * **Tools not appearing:** Check JSON syntax and restart client
    * **Connection errors:** Verify Docker is running or Python path is correct
    * **SEC errors:** Ensure User-Agent includes valid email
    * **Rate limiting:** Add delays between requests
  </Accordion>
</AccordionGroup>

## Finding More Clients

The MCP ecosystem is rapidly growing. Find the latest clients and tools:

* [MCP Ecosystem Directory](https://modelcontextprotocol.io/ecosystem)
* [Awesome MCP Servers](https://github.com/punkpeye/awesome-mcp-servers)
* [MCP GitHub Topics](https://github.com/topics/mcp)
* [MCP Discord Community](https://discord.gg/modelcontextprotocol)

## Next Steps

<CardGroup cols={2}>
  <Card title="Configuration Guide" icon="gear" href="/setup/configuration">
    Advanced configuration options
  </Card>

  <Card title="Docker Setup" icon="docker" href="/setup/docker">
    Detailed Docker installation guide
  </Card>

  <Card title="Troubleshooting" icon="bug" href="/setup/troubleshooting">
    Common issues and solutions
  </Card>

  <Card title="Tools Overview" icon="wrench" href="/tools/overview">
    Explore available SEC Edgar tools
  </Card>
</CardGroup>