> ## Documentation Index
> Fetch the complete documentation index at: https://docs.abundly.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Atlassian

> Work with Jira, Confluence, and other Atlassian products

Connect your agent to Atlassian to work across Jira, Confluence, and other Atlassian Cloud products — search issues, create pages, manage projects, and more.

## How to enable

The recommended route is the Atlassian MCP server (formerly Rovo MCP server): an Atlassian organization admin must first allow Abundly to access your Atlassian instance, after which each user authenticates with their own Atlassian account. If your admin can't allow the domain, connect [with an API token](#connecting-with-an-api-token-instead) instead.

**1. Allow Abundly in your Atlassian admin console**

1. Go to [admin.atlassian.com](https://admin.atlassian.com) and select your organization
2. Navigate to **Connectors → Atlassian MCP server** (in some organizations: **Rovo → Rovo MCP server → Domains**)
3. Click **Add domain** and enter your Abundly app domain with `/**` (for example, `https://app.abundly.ai/**` or `https://your-tenant.app.abundly.ai/**`)

![Atlassian Administration showing Connectors → Atlassian MCP server, Domains tab, with https://app.abundly.ai/\*\* added under Your domains](https://mintcdn.com/abundly/Bx3aQ6nW7lKqvCuE/images/atlassian-mcp-server-domains.png?fit=max\&auto=format\&n=Bx3aQ6nW7lKqvCuE\&q=85\&s=154ff54a9dd2670373e21887b447c915)

This is a one-time setup. Once the domain is added, all users in the organization can connect their accounts.

**2. Choose what agents may do**

On the **Permissions** tab of the same page, allow **Read**, **Write**, and **Search** per Atlassian app (Jira, Confluence, and others). Match these to what your agents should do — an agent that should create Jira issues needs Write on Jira, while a reporting agent only needs Read and Search.

**3. Check your data security policies**

If your organization uses data security policies, go to **Security → Data protection → Data security policies** and make sure the **Atlassian MCP server** control is set to **Allowed** for the content your agents need. When it's blocked, agents can't read the covered Jira issues or Confluence pages, even though the connection itself works.

**4. Add Atlassian to your agent**

Go to the **MCP Servers** tab in your agent's capabilities. You'll find Atlassian in the "Add a server" section — click it to add.

**5. Authenticate your account**

Click **Connect**, then **Continue with OAuth** and complete the OAuth flow to connect your Atlassian account.

> [!NOTE]
>
> The server lists only around 20 common tools, even though Atlassian offers hundreds of operations across Jira, Confluence, Loom, Goals, and more. For anything else, the agent uses the `discover` tool to find the right operation and then calls it, so you don't need to enable more tools.

> [!TIP]
>
> API token authentication is also available under advanced settings, but OAuth is recommended — API tokens only expose a limited subset of tools. API token authentication is off by default: an admin must turn it on under **Connectors → Atlassian MCP server → Authentication → API token**. The domain allowlist from step 1 doesn't apply to API tokens. To create a personal API token or a service account API key and use it with the MCP server, see Atlassian's guide to [configuring authentication via API token](https://support.atlassian.com/atlassian-rovo-mcp-server/docs/configuring-authentication-via-api-token/).

## Troubleshooting

* **Calls fail even though the connection succeeded** — if your Atlassian organization restricts access with an IP allowlist, add [Abundly's outbound IP addresses](https://docs.abundly.ai/integrations/github.md#ip-allowlisting) to it. Atlassian otherwise rejects the agent's calls, even after OAuth succeeds.
* **Authentication or connection problems** — remove the Atlassian server from the agent in Abundly, add it again, and complete the OAuth flow once more.

## Connecting with an API token instead

The Atlassian MCP server is the easiest route, but it isn't the only one. You can also give an agent direct access to the Jira and Confluence REST APIs with an Atlassian API token, defined as an [Outbound API](https://docs.abundly.ai/features/http.md#handling-credentials). This is worth considering when:

* Your Atlassian admin can't or won't allow the Abundly domain in the Atlassian MCP server settings
* The agent should act as its own dedicated Atlassian account rather than on behalf of a person
* You need endpoints or fields the MCP server doesn't expose

The trade-off is that you describe the API to the agent yourself instead of getting a ready-made tool set.

**1. Create the API token**

Go to [id.atlassian.com/manage-profile/security/api-tokens](https://id.atlassian.com/manage-profile/security/api-tokens) while signed in as the account the agent should act as, and create either a classic token or a token with scopes (see below). Copy the value immediately — it's only shown once.

**2. Store it as a Secret**

In **Workspace → Secrets** (or straight in the keyhole when you define the Outbound API in the next step), create a secret whose value is the Atlassian account email and the token joined by a colon:

```
agent@yourcompany.com:your-api-token
```

Store it unencoded — the platform Base64-encodes it and adds the `Basic ` prefix when calling the API.

**3. Define the Outbound API**

From the agent's Capabilities page (or **Workspace → Outbound APIs**), create an outbound API with a **Basic auth** keyhole and the endpoint pattern for your token type, and put the secret in the keyhole — or paste the `email:token` value into the keyhole to create the secret in the same step. Created from the agent, the API is enabled on it right away; otherwise enable it on the agent's Capabilities page. The general [HTTP Requests capability](https://docs.abundly.ai/features/http.md) is not required — an Outbound API works on its own.

### Classic vs. scoped tokens

Atlassian offers two kinds of API token, and they are called against **different base URLs** — this is the detail most setups get wrong.

| | Classic token | Token with scopes |
| - | - | - |
| **Permissions** | Everything the account can do | Only the scopes you select |
| **Base URL** | `https://your-site.atlassian.net` | `https://api.atlassian.com/ex/jira/{cloudId}` |
| **Endpoint pattern** | `https://your-site.atlassian.net/rest/` | `https://api.atlassian.com/ex/jira/` |
| **Example call** | `/rest/api/3/search/jql` | `/rest/api/3/search/jql` (after the base URL) |

Prefer a scoped token: it limits the damage a leaked credential can do, and it lets you give the agent read-only access where that's enough. Pick the smallest set of scopes that covers the agent's job — read-only on issues is enough for reporting and triage suggestions, and you only need write scopes if the agent should change anything. Confluence uses the same pattern with `ex/confluence` in place of `ex/jira`.

To find your `cloudId`, open `https://your-site.atlassian.net/_edge/tenant_info` in a browser — or simply ask the agent to fetch it.

> [!WARNING]
>
> Atlassian API tokens expire — up to 365 days, and one year by default. Put a reminder in your calendar, or have the agent remind you: when the token expires, its calls start failing with authentication errors. Rotating the token means updating the Secret; the Outbound API and the agent stay untouched.

Once the agent has a working call, ask it to write the details into its [instructions](https://docs.abundly.ai/features/instructions.md) — base URL, JQL it used, fields it needs. That turns trial and error into a repeatable routine.

## Example use cases

* **Issue triage** — "Every morning, check Jira for new unassigned issues in the Support project and suggest an assignee based on workload"
* **Sprint reporting** — "Summarize the current sprint progress and post an update to Slack every Friday"
* **Documentation sync** — "When a Jira epic is marked as done, create a Confluence page documenting what was shipped"
* **Cross-platform updates** — "When someone mentions a Jira ticket in Slack, look it up and post a summary with the current status"


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.