How to Connect Atlassian (Jira & Confluence) to Cursor — via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Atlassian (Jira, Confluence), paste your credentials, mint an MCP API key — done. No Docker, no git clone, no local server.

Atlassian (Jira & Confluence) + Cursor

Work with Jira and Confluence from Claude, ChatGPT or Copilot through Atlassian's official Rovo MCP server: search and update work items, read and write pages, run JQL and CQL, and ask Rovo search across your Atlassian apps.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

ATLASSIAN_EMAIL, ATLASSIAN_API_TOKEN

Step 1 — Get credentials

AnythingMCP connects to the Atlassian Rovo MCP Server (https://mcp.atlassian.com/v2/mcp), Atlassian's official remote MCP server, with your Atlassian email and a scoped API token. It covers Jira (work items, JQL, boards and sprints), Confluence (pages, spaces and CQL), Rovo search across your Atlassian apps and, where your organization has them, Jira Service Management, Bitbucket Cloud, Loom and Goals. At install AnythingMCP reads the tools Atlassian offers your token, each with the prefix atlassian_. When Atlassian adds tools or your admin enables another permission group, open the connector and click Discover tools.

1. Admin step: an Atlassian organization admin turns on API-token authentication under Atlassian Administration → Rovo → Rovo MCP server settings. Without it every call is refused. The same settings control the permission groups; delete_jira and manage_jira are off by default.
2. Signed in as the person the AI should act as, open id.atlassian.com/manage-profile/security/api-tokens and choose Create API token with scopes. Pick the agent-interface scopes of the apps you need, for example read:jira:agent-interface, write:jira:agent-interface, search:jira:agent-interface, read:confluence:agent-interface, write:confluence:agent-interface, search:confluence:agent-interface and search:rovo:agent-interface. Copy the token right away: Atlassian shows it only once.
3. Set ATLASSIAN_EMAIL to the email of the account that owns the token, and ATLASSIAN_API_TOKEN to the token (it usually starts with ATATT). Install. AnythingMCP sends both as HTTP Basic auth, which is how Atlassian accepts personal API tokens.
4. Service account instead of a person: an admin creates a service account and an API key with the same scopes. Atlassian takes that key as a Bearer token, so after installing open the connector editor, change the auth type to Bearer token and paste the key there.

IP allow list: if your organization restricts access with an IP allow list, add the address AnythingMCP calls from (on AnythingMCP Cloud, ask support for the current egress IP). With API tokens the IP allow list is the only network check.

Switched off at install: atlassian_executeWrite and atlassian_executeDestructive (they run any operation of Atlassian's catalogue by name, so atlassian_executeWrite alone could merge a pull request or change permissions), the Jira delete tools, Jira project and screen administration, sprint and version management, entity properties, Confluence permission, restriction, public-link and space-instruction changes, the Confluence Remix tools (15 to 30 Rovo credits each), Bitbucket merge, approve, commit and pipeline runs, Loom visibility changes and Jira Service Management alert actions. Switch any of them on per tool when you need it. atlassian_search and the Teamwork Graph tools use up to 10 Rovo credits per call from your organization's shared pool. Every call runs with the token owner's Jira, Confluence and Bitbucket permissions, so use an account that sees only what the AI should.

First call: atlassian_getAccessibleAtlassianResources. API tokens are not bound to one site, so almost every tool needs a cloudId, and this call returns it for each site you can reach. Then use atlassian_search for questions in plain language, atlassian_searchJiraIssuesUsingJql or atlassian_searchConfluence for exact JQL and CQL filters, and atlassian_getJiraIssue or atlassian_getConfluenceContent to read one item. Code search, Teams and some Compass tools need OAuth 2.1 and do not work with an API token. A 401 means API-token authentication is off for your organization, or the email and token do not belong together; a scope error on one tool means the token lacks that app's agent-interface scope, so create a new token with it and replace it in the connector.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Cursor

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Cursor reads MCP servers from ~/.cursor/mcp.json. Add this entry:

{
  "mcpServers": {
    "anythingmcp": {
      "url": "https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID",
      "headers": { "X-API-Key": "YOUR_MCP_API_KEY" }
    }
  }
}

1. Create an MCP API key in AnythingMCP under MCP Servers → your server → API keys, and paste it in place of YOUR_MCP_API_KEY.
2. Save the file and restart Cursor.
3. Open Cursor → Settings → MCP and check that anythingmcp is listed and connected.
4. Start chatting — the agent can now call the Atlassian (Jira & Confluence) tools.

Available tools

(showing 30 of 193 tools — install the connector to see them all)

FAQ

Does Cursor support custom MCP servers on the free tier? Yes — MCP is available on Hobby, Pro and Business plans.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Gemini
• Connect to Meta Muse