Technology

This section covers how the Corridor platform is installed, configured, and operated. It is written for the platform and infrastructure teams who run Corridor, rather than for the analysts who use it day to day.

For guidance on using the platform itself, see the User Guide.

AI

How CORRA AI and Corridor MCP work, and how to connect an external agent.

• AI Architecture: what CORRA AI and Corridor MCP are, and how the server runs
• Bringing Your Own LLM: provider and credential configuration
• Installing Corridor MCP: hosting and client setup
• What Corridor MCP Can Do: the available capabilities, and the limits
• Input and Output Guardrails: how access is controlled and how the model’s inputs and actions are constrained
• Observability and Traceability: logs, conversation records, and model telemetry

Architecture

Understand how the platform is put together before installing it.

• Architecture Overview: the analytical layer and the production layer
• Components: the services that make up an installation
• Networking: how those services communicate
• Scalability and Sizing: where to add resources when things slow down

Installation

Everything needed to stand up a new instance.

• Minimum Requirements
• Downloading the Platform
• Bundle Install and Docker Install
• Configuration Setup: API, web application, and Jupyter
• Components: web server, metadata database, authentication, file management, process management, and data lake integration

Customization

Adapt the platform to your environment and processes.

• Customization Overview

Maintenance

Keep a running instance healthy.

• Maintenance Overview
• Upgrading Corridor
• Backups
• Securing the Platform

Production

Run approved logic outside the analytical layer.

• Production Overview

CLI: Utility Commands

Command line tools for operating the platform.

• CLI Overview