Skip to main content

​
Overview

MCP (Model Context Protocol) allows AI tools to connect directly to your context repository. Once connected, AI agents can retrieve, search, and update context while respecting the permissions and access controls configured in Qontext. Examples include:

​
Connecting via MCP

Qontext exposes a hosted MCP server: 
https://api.qontext.ai/mcp
Configure this URL in your MCP client and complete the authentication flow. Please refer to the Clients section for client-specific setup instructions.

​
Available operations

Depending on the permissions granted to the client, MCP can be used to:
  • Browse files and folders
  • Read files
  • Search the context repository
  • Follow references between files
  • Create files
  • Update files
  • Manage folders
This allows agents to use the context repository as both a source of context and a workspace for creating and maintaining context.

​
Access controls

Each MCP client can have its own:
  • System instructions
  • Access controls
  • Permissions
This allows different agents to access different parts of the context repository. Please also refer to system instructions and permissions.