VE

VeyraX MCP Server: Centralized Tool Control

Control all your tools from one centralized connection with the VeyraX MCP server, accessing every MCP-compatible environment seamlessly.

Quick Install
npx -y @VeyraX/veyrax-mcp

Overview

VeyraX MCP Server acts as a single, centralized bridge between your tools and any environment that supports the Model Context Protocol (MCP). Instead of wiring each tool into every assistant, editor, or IDE you use, run one MCP server that exposes all your VeyraX-connected tools to any MCP-compatible client. The server handles authentication and tool discovery so client integrations remain simple and consistent.

For developers building integrations or operators running internal tool hubs, VeyraX MCP reduces duplicated configuration, removes repeated authentication flows, and lets clients uniformly discover and call tools over the MCP surface. This is particularly useful when you want the same set of tools available in multiple environments (e.g., Claude, Cursor, VS Code) without per-client credential management.

Features

  • Centralized authentication to VeyraX: connect once and serve tools to many clients
  • Standard MCP API surface for tool discovery and invocation
  • Lightweight server you can run locally, in a container, or as a managed service
  • Quick setup: sample configs and baked-in defaults for common MCP clients
  • Compatible with MCP-capable environments: Cursor, Claude, Windsurf, VS Code (Cline)
  • Health and diagnostics endpoints for monitoring and troubleshooting

Installation / Configuration

Clone the repository and run locally, or use the provided Docker image.

Install and run locally (Node.js example)

git clone https://github.com/VeyraX/veyrax-mcp.git
cd veyrax-mcp
# Install dependencies (example: npm)
npm install
# Set your VeyraX API key (obtain from https://www.veyrax.com/register)
export VEYRAX_API_KEY="sk_XXXXXXXXXXXXXXXX"
# Start server (default port 3000)
npm start

Run with Docker

# Build
docker build -t veyrax-mcp .

# Run (adjust port and API key)
docker run -e VEYRAX_API_KEY="sk_XXXXXXXXXXXXXXXX" -p 3000:3000 veyrax-mcp

Example environment variables and options

VEYRAX_API_KEY=sk_XXXXXXXXXXXXXXXX   # required: your VeyraX API key
PORT=3000                            # optional: port to listen on
MCP_BASE_PATH=/mcp                   # optional: base path for MCP endpoints
LOG_LEVEL=info                       # optional: debug/info/error

Typical MCP client configuration (paste into client settings)

{
  "name": "VeyraX MCP",
  "mcp_url": "http://localhost:3000/mcp",
  "api_key": "sk_XXXXXXXXXXXXXXXX"
}

Quick health checks

# Check server health
curl http://localhost:3000/health

# List available tools (MCP discovery)
curl -H "Authorization: Bearer sk_XXXXXXXXXXXXXXXX" \
     http://localhost:3000/mcp/tools

Available Resources

  • GitHub repository: https://github.com/VeyraX/veyrax-mcp
  • Documentation: https://docs.veyrax.com/mcp
  • Support: [email protected]

Supported MCP clients (examples)

ClientNotes
CursorIDE integration; paste MCP URL into Cursor settings
ClaudeUse MCP endpoint in agent integrations
WindsurfIDE support for tool invocation
VS Code (Cline)Integrate via Cline extension settings

Use Cases

  1. Unified developer tools in multiple UIs
    Run VeyraX MCP internally and expose your collection of tools (code search, deploy triggers, secret-aware snippets). Developers can access the same tools from Cursor, VS Code, or Claude without re-authenticating per client.

  2. Secure internal tools behind a single auth boundary
    Instead of distributing credentials for internal utilities, store credentials centrally in VeyraX and let the MCP server mediate access. Clients receive tool metadata and invoke actions while the server enforces access controls.

  3. Rapid prototyping with assistant-driven workflows
    Attach debug or analysis tools (log search, test-runner, metrics query) to VeyraX and use them inside an LLM-based assistant for interactive debugging sessions. The MCP server provides a predictable API so assistant prompts can call tools programmatically.

  4. CI / automation invocations from editors
    Trigger build or deployment steps from within an editor (e.g., VS Code) by invoking MCP tools that wrap your CI pipeline endpoints. This keeps pipeline credentials centralized and audit-ready.

Best Practices & Troubleshooting

  • Ensure VEYRAX_API_KEY is correct and not publicly exposed. Use secrets management in production.
  • Configure TLS (HTTPS) and firewall rules when exposing MCP endpoints beyond localhost.
  • If a client cannot discover tools, verify CORS and base path settings, and inspect server logs for authentication failures.
  • Use the /health and /metrics endpoints to integrate with monitoring systems.

For full API details, client-specific guides, and advanced configuration, see the official docs at https://docs.veyrax.com/mcp and the project repo at https://github.com/VeyraX/veyrax-mcp.