Model Context Protocol (MCP) Connectivity - Axiooma

Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open standard that allows secure connections between AI language models and local software tools or data sources. Axiooma features native support for MCP, enabling AI agents to read context from your local development environment to refine and enrich software specifications in real time.

Why Use MCP in Requirements Engineering?

Standard cloud-based AI tools operate in isolation from your local environment. They cannot access local file folders, database schemas, or project APIs. MCP bridges this gap:

• Database Aware: The AI can query database schema configurations to write consistent validation logic.
• Filesystem Context: Allows reading local API specs, type definitions, or config parameters to check dependencies.
• Secure by Design: The MCP server runs locally on your machine. You specify which files, tables, or APIs are exposed to the Axiooma AI agents.

Supported MCP Servers

You can connect any standard MCP server to the workspace:

1. Filesystem Server: Grants the AI read-only access to selected directories in your project.
2. Postgres/MySQL DB Reader: Enables the AI to query schema structures to guarantee database compatibility.
3. Microsoft 365 Suite: Fetches and references project meeting notes, emails, or tasks as business context.

Connecting an MCP Server in Axiooma

1. Go to the Settings panel from the sidebar.
2. Click on the MCP tab.
3. You will see a list of configured MCP server connections for your organization.
4. To add a server, enter the connection credentials or environment keys and click Authorize & Save.
5. Click Test Connection to verify that the local communication channel is online.

Once configured, Axiooma’s AI Assistant automatically leverages the tools and databases exposed by your local MCP server during requirement writing and refinement workflows.