There is no single approval setting that governs every AI agent in Atlassian. For a Rovo agent, control who can create and use it in Rovo Studio, select whose identity it acts under, and limit its app and content access and available tools. Interactive Rovo agents request confirmation before certain consequential actions; agents running in automations may act without a person reviewing each action. Atlassian MCP and third-party Agent2Agent (A2A) connections have separate organization-level controls.
First identify which kind of agent access you need to control
These settings apply to different ways of using agents, so start with the relevant control surface rather than treating “AI agent permissions” as one switch.
As an Amazon Associate I earn from qualifying purchases.
- Rovo agent in an interactive conversation: Set agent access, identity, and tools in Rovo Studio. Interactive confirmation applies to documented consequential tools that may mutate data across systems.
- Rovo agent in an automation: Review the identity used by the flow and whether agents are allowed to act in automations. Do not assume a person will confirm every action.
- Atlassian MCP server: Set its Read, Write, and Search permissions in Atlassian Administration.
- Third-party A2A connection: A security or organization administrator controls the organization-wide A2A toggle; the connecting user still needs authorization and existing app access.
The Rovo Studio creation-setting availability described in Atlassian Support’s “Rovo agent permissions and governance” is listed for Atlassian Cloud Standard, Premium, and Enterprise; Government Cloud is not available for that setting. Confirm current availability and labels in your tenant, as product settings can change.
Control who can create, edit, manage, and use Rovo agents
Restrict agent creation
- Open Rovo Studio and select Settings as a Studio admin.
- Under agent creation, choose who may create agents. The default is All users. Choose Selected groups and add up to 10 user groups, or choose No users to restrict creation to the admin group.
This creation rule controls who can make agents; it does not by itself determine who can edit or use an existing agent.
Assign editors and managers, then set visibility
In an agent’s Users and permissions settings, the owner can add people as editors or managers. Editors can edit the agent. Managers can edit it, add other editors, and delete it. Agent visibility is open to everyone by default. To limit access, turn Open to all users off and add permitted people individually, assigning an editor or manager role as appropriate. Atlassian’s current guidance says visibility restrictions cannot be assigned to groups or teams, so this control requires individual additions.
Choose the identity the agent will use and scope its access
In the agent’s Access and identity settings, choose User’s account or Agent’s account. That choice determines whose permissions the agent can use and how its work is attributed; it does not let the agent exceed the selected identity’s permissions.
| Identity | Permissions used | How work is attributed | Documented fit |
|---|---|---|---|
| User’s account | The interacting user’s permissions. In an automation, this can mean the account of the person who created the flow. | Under that user | Interactive or personal assistance. Review carefully before using in automation if the flow creator has broad access. |
| Agent’s account | A separately managed identity whose access is set by organization, app, space, or content administrators. | Under the agent | Automation that should use a distinct, limited identity rather than depend on an individual’s credentials. |
Whichever identity you choose, check both the relevant organization or app access and the specific spaces, pages, or other content the agent needs. Access to an app does not automatically mean the identity can read every space or page. Atlassian describes the agent account as the better fit where automation should not rely on a user’s credentials; see its Support guides “Understand Rovo agent accounts” and “Configure a Rovo agent’s identity.”
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Limit the tools available to an interactive Rovo agent
Give the agent only the tools required for its task. The tools determine which actions are available; instructions can describe intended limits, but should not substitute for restricting unnecessary capabilities. Atlassian Support’s “Add tools to Rovo agents” says: “The agent will respond asking for confirmation before executing consequential tools that may mutate data across systems.” This is a documented interactive confirmation behavior, not evidence of a universal, administrator-configurable approval matrix for every agent action.
Handle automation approvals as a separate workflow decision
An agent used in an automation may act without a person confirming each action. Atlassian’s “Best practices to automate agents safely” states: “In automations, there is no user to interact with, review, or approve an action.” Treat automation as autonomous unless you deliberately add a human-review step to the surrounding workflow.
Reduce the risk of automated actions
- Where possible, use the agent’s own account and grant it only the access required for the flow.
- Use the read-only setting in the Use agent step when the automation needs an answer but should not let the agent perform write actions.
- Administrators and users can prevent agents from acting in automations. If agent actions are blocked, write tools fail; the flow can still use the text response through
{{agentResponse}}in later actions. - If a consequential change needs a person’s decision, add an explicit review step in the workflow rather than assuming the interactive Rovo confirmation will appear during automation.
These behaviors are described in Atlassian Support’s “Automate Rovo agents” and “Best practices to automate agents safely.”
Set permissions for the Atlassian MCP server
MCP server access is separate from an agent’s Rovo Studio tool settings. An organization admin can review it at Atlassian Administration > Rovo > Rovo MCP server > Permissions.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problems- Review the Read, Write, and Search controls against the actions your organization intends to allow.
- Select Edit details to set permissions for individual apps.
- Decide whether those app permissions should also be applied automatically to future additions.
Atlassian Support’s “Configure Atlassian MCP server permissions” says these MCP controls take precedence over Connected Apps or individual Marketplace app settings for MCP access. They do not replace the other Rovo agent or content-access controls.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Review third-party Agent2Agent access separately
A2A is disabled by default. Before enabling it, complete the organization’s security and compliance review. An organization admin can then go to Atlassian Administration > Rovo > Agent2Agent and enable Allow A2A.
The setting is organization-wide and cannot be scoped per Atlassian app. Enabling it does not grant a third-party agent access that the user or app does not already have: the connection still requires existing app access and valid OAuth 2.1 user authorization. These controls are described in Atlassian Support’s “Manage Agent2Agent connections.”
Include connected apps and Rovo AI activation in the review
Before connecting an external source, check its permissions. Atlassian’s AI Trust guidance says admin-managed connectors are not enabled by default and respect existing user permissions. Organization admins can manage activation of Rovo AI-powered features by app. Some non-AI Rovo features are part of the platform and cannot be disabled, so turning off AI-powered features should not be treated as disabling every Rovo capability.
What to check before putting an agent into use
- Identify whether the agent is interactive, part of an automation, using MCP, or connected through A2A.
- Limit creation, editing, management, and visibility separately, according to who needs each capability.
- Choose an identity deliberately and confirm its app and content permissions are no broader than needed.
- Restrict tools and write access; for automation, add a human-review step if a person must approve a decision.
- Check connector permissions and the relevant MCP or A2A controls rather than assuming general Rovo settings cover them.
Menu labels, roles, and availability can change. The steps above reflect Atlassian’s public guidance checked on October 7, 2026; verify the current controls in your Atlassian Cloud tenant.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
