Skip to main content
Connect Jira and Confluence when customer conversations need to become, or be checked against, engineering and documentation work. The Rippit agent can find the Jira issue a run of tickets is already tracking, file a new one with the evidence attached, or read the Confluence page that states the policy an agent should have followed. This connector uses Atlassian’s hosted Rovo MCP server and covers Atlassian Cloud only.

​
What the agent gets

What you get depends on how you connect, so read Two ways to connect before choosing.
  • Jira issues and projects: search with JQL or text, read issues and comments, and create or update issues and comments when write tools are enabled.
  • Confluence pages and spaces: search and read pages, and create or update them when write tools are enabled.
  • Teamwork Graph context: for any Jira issue, Confluence page, or other Atlassian object, everything Atlassian has linked to it, such as related issues, pages, pull requests, deployments, and people, and the full content of those linked objects.

​
Two ways to connect

​
Your Atlassian account (OAuth)

Click connect, sign in to Atlassian, pick the site, and approve. There is no Atlassian app to create first, because Rippit registers itself as an OAuth client automatically. The connection acts as you and sees what you can see in Jira and Confluence. This is the path that exposes Atlassian’s Jira and Confluence tools. It is the right choice for anything interactive.
Atlassian admins can restrict which domains may connect to the Rovo MCP server, under Atlassian Administration → Rovo → Rovo MCP server → Domains. If sign-in fails with a message about an unapproved client or domain, ask your Atlassian admin to add app.rippit.com there.

​
A service-account API key

Paste an API key that an Atlassian admin created for a service account. The connection then acts as that service account, is organization-wide by nature, and does not break when a person leaves. This is the better fit for scheduled Agent Apps. Two things to know before choosing it:
  1. An Atlassian admin has to enable it. API-key authentication to the MCP server is off by default. The admin turns it on under Atlassian Administration → Rovo → Rovo MCP server → Authentication, creates the service account, and generates a key with the scopes the agent needs.
  2. The key’s scopes decide which tools exist. Atlassian filters the tool list by the scopes on the credential. A key without Jira and Confluence read scopes exposes only the Teamwork Graph tools listed below, which is what the API-key connections we have seen so far look like. Ask the admin to grant the Jira and Confluence scopes the agent needs when the key is created, and check Manage tools after connecting to see what arrived.

​
Identity and scope

An OAuth connection acts as the person who authorized it and inherits their Jira and Confluence permissions, so an issue or page that person cannot open does not become readable through Rippit. An organization-wide OAuth connection extends that person’s access to everyone using the agent; authorize with an account whose access you are comfortable sharing that widely. An API-key connection acts as the service account, with exactly the access the admin granted it. The scope you choose at setup decides whether Agent Apps can use the connection, exactly as with every other connector.

​
Tools

Atlassian’s tool list varies with the sign-in path and the credential’s scopes, and Atlassian is moving its server to a new version that changes the list again, so this page does not enumerate the Jira and Confluence tools. Manage tools on the connection card is the authoritative list for your connection. The Teamwork Graph tools below are present on every connection, including API-key connections with no Jira or Confluence scopes. The read tools are enabled by default; the write tool stays off until a Rippit Full Admin enables it. Atlassian’s own documentation for the server is at Atlassian Rovo MCP Server.