MCP Connectors

Connectors are managed connections to MCP servers that you set up in JetBrains Air Teams. Connecting a service makes its tools available to your cloud tasks and automations, so the agent can fetch data and perform actions in that service.

JetBrains Air Teams has connectors for:

Confluence – pages, spaces, and attachments
Figma – files, comments, and design resources
GitHub – repositories, pull requests, and issues. Always on, see GitHub connector.
GitLab – repositories, merge requests, and issues
Grafana – dashboards, alerts, and data sources
Jira – projects and issues
Linear – projects and issues
Sentry – errors and application health
TeamCity – builds, test results, and investigations
YouTrack – issues, agile boards, and articles

GitHub connector

The GitHub connector needs no setup and can't be disabled. It shows under Automatic on the Connectors page, without a Connect button, and the GitHub MCP server is available to the agent in every cloud task and automation run that works on a GitHub repository.

Personal and project connectors

You can add connectors at two levels:

Personal – available in your own cloud tasks and automations. You add them in Settings | Connectors and sign in with your own account.
Project – available to every cloud task and automation in a project. A project admin adds them on the project page and signs in with a shared account, so all project members use the same connection.

Add a personal connector

Connect a service

Open JetBrains Air Teams at https://air.jetbrains.cloud.
Open Settings and go to Settings | Connectors.
Find the service you want under Available. To filter the list, type in Search connectors.
Click Connect and follow the provider's authorization flow to sign in and grant access. Services that run on your own server also ask for its address.
JetBrains Air Teams reaches your server from its own connector service, not from the task's run environment, so the address has to be reachable from the internet. Services that are only available on an internal network or behind a VPN can't be connected this way. Add it through a mcp.json file instead, and run the task in an environment that can reach it.
When you finish, the service moves to the Connected list. You can then select the connector in your cloud tasks and, through Select MCP connectors, in your automations.

Add a connector to a project

Project admins add connectors on the project page. A project connector is available to every cloud task and automation in the project, so you set it up once instead of each member configuring the same MCP server.

For details, see Projects | Add connectors to a project.

MCP servers from a mcp.json file

A connector is the easier of two ways to give an agent MCP tools: you pick a service, sign in, and JetBrains Air Teams handles the connection. The other way is a mcp.json file, which you write yourself.

Reach for mcp.json when a connector can't do the job:

the service isn't in the connector list
the server is only reachable from inside your network, so the connector service can't sign in to it, but the task's run environment can
you want the configuration versioned with the code, so it changes through review like anything else in the repository

Configure a server for cloud tasks

A cloud environment has no settings screen, so its MCP configuration comes from the repository. Put a .mcp.json file in the repository root, and the environment picks it up with the checkout.

Don't write tokens into the file. Reference an environment variable with ${VARIABLE} instead, and supply the value as a secret in the environment configuration, so it stays out of the repository:

                {
                    "mcpServers": {
                        "youtrack": {
                            "url": "https://youtrack.example.com/mcp",
                            "headers": {
                                "Authorization": "Bearer ${YOUTRACK_MCP_TOKEN}"
                            }
                        }
                    }
                }
            

For the Air app, which configures MCP servers through its own settings as well as through files in the project, see MCP servers.

How the two combine

Connectors and mcp.json servers work together. If a repository has both, the agent gets the tools from all of them at the same time.

Set up each service one way or the other, though. If the same service comes from both a connector and mcp.json, the agent gets two copies of the same tools and can pick the wrong one.

30 September 2026