--- title: "Jira and Confluence (Atlassian)" description: "Let a playbook search and read Jira work items and Confluence pages, and create, update and comment on them, through Atlassian's MCP server, as you." url: "https://on.auto/integrations/atlassian" last_updated: "2026-11-10T08:19:38Z" --- # Jira and Confluence (Atlassian) Search and read work items and pages, and create, update and comment on them. An Atlassian connection gives a playbook's functions Atlassian's MCP server, with every tool it lists for Jira and Confluence. You sign in to Atlassian, choose a site and allow access, and the connection acts as you, with your permissions on that site. Available on Auto Cloud and on self-hosted runtimes. * * * ## What a playbook can do with Jira and Confluence * Search Jira work items with JQL, and read work items, projects, boards, sprints, comments, worklogs and attachments. * Create and change work items, comments, worklogs, links, versions, sprints and boards. * Search Confluence with CQL, and read pages, spaces, comments, attachments and versions. * Create, change, move and archive pages, spaces, comments and labels. The connection asks none of the scopes that delete or administer in Jira, so a playbook cannot delete a work item or change a project's settings. Atlassian's MCP server lists its own tools when a function runs, and once you connect, your agent lists them with `list_connection_tools`. * * * ## An example workflow Triage of new bugs, the way the team lead does it. A schedule trigger runs the workflow every hour. A lookup function runs the JQL `project = APP AND type = Bug AND created >= -1h`. A reasoning function reads each bug with the team's triage guide in its prompt: the components and their owners, what counts as a blocker, the questions a good report answers, and decides a priority, a component and an owner, or that the report needs more. An interaction function updates the work item and comments with the reasoning, or comments with the questions the reporter should answer. For a bug it rates a blocker, a second interaction function asks the team lead in Slack and waits for a confirmation before it sets the priority. A Confluence lookup function reads the release page so that a bug against the current release is labelled as such. The run's history in Studio keeps every decision and its reason. * * * ## How to connect Jira and Confluence (Atlassian) * **Sign in to Atlassian, choose a site and allow Jira and Confluence.** The consent screen shows the permissions listed below. Then: 1. Ask your agent to connect Jira and Confluence (Atlassian) to Auto, or open **Settings**, **Agents**, **Connections** in [Auto Studio](https://studio.on.auto/) and choose Atlassian. 2. On the Studio page, check the service and its access, choose which playbooks may use the connection, and approve it. 3. Have your agent check one small read through the connection before a function depends on it. * * * ## What it needs * An Atlassian account on the site, whose Jira and Confluence permissions the connection has. * An Atlassian organization admin who allows the address people return to, where your organization restricts apps, as [Let people add connections](https://on.auto/docs/self-host/connections) notes for Atlassian. * Your agent connected to Auto and signed in as you, with permission to change your organization, if your agent adds the connection. * On a self-hosted runtime, nothing to register: Atlassian's authorization server takes a client Auto names by its metadata document or registers itself, or one you registered. ### Permissions Auto asks for * Read the Atlassian profile of the person who connects. * Read the profiles of Atlassian users. * Read the email address of the Atlassian account that connects. * Keep access to Jira and Confluence when nobody is signed in. * Read Jira work items, projects, boards, sprints, comments, worklogs and attachments. * Search Jira work items with JQL. * Read Confluence pages, spaces, comments, attachments and versions. * Search Confluence content with CQL. * Create and change Jira work items, comments, worklogs, links, versions, sprints and boards. * Create, change, move and archive Confluence pages, spaces, comments, labels and permissions. * * * ## Good to know A tool that creates or changes is marked as one that acts, so a lookup function cannot call it. What a function reads, a work item's description or a page's text, is kept in the run's history, where people with access to the playbook can read it. * * * [All integrations](https://on.auto/integrations/) · [Linear](https://on.auto/integrations/linear) · [GitHub](https://on.auto/integrations/github) · [Notion](https://on.auto/integrations/notion) · [Bring in business data](https://on.auto/docs/cloud/connections) · [Give a function tools](https://on.auto/docs/guides/tools) · [Pricing](https://on.auto/pricing) · [Atlassian documentation](https://support.atlassian.com/atlassian-ai-gateway/docs/get-started-with-the-atlassian-remote-mcp-server/) ## Sitemap See the [full documentation and product sitemap](https://on.auto/sitemap.md) for related pages.