All integrations
Atlassian (Jira and Confluence)
Jira issues and Confluence pages across your Atlassian sites.
Productivity
Developer
Docs and knowledge
How an agent reaches it
- MCP
Official MCP server. The agent lists its tools and calls them.
- REST API
Always available. The agent calls the API directly with your credential attached.
- CLI
No official command-line tool.
How you authenticate
- OAuth
- Username and password
- Create an API token at id.atlassian.com, then Security, then API tokens.
- Your account email is the Basic username and the token is the password.
How agents reach it
- Base URL
- https://api.atlassian.com/ex/jira/{cloudid}/rest/api/3
- Pagination
- offset
- MCP server
- https://mcp.atlassian.com/v1/mcp
Common endpoints
| discover sites (do this first) | GET https://api.atlassian.com/oauth/token/accessible-resources |
| search issues | GET /search?jql={jql} |
| read an issue | GET /issue/{issueIdOrKey} |
| create an issue | POST /issue |
| list Confluence pages | GET https://api.atlassian.com/ex/confluence/{cloudid}/wiki/api/v2/pages |
Worth knowing
- The base URL is per-SITE and unusable until cloudid is discovered. This is the single most common way an Atlassian call fails.
- Jira and Confluence have separate base paths under the same cloudid: /rest/api/3 and /wiki/api/v2.
- JQL in a query string must be URL-encoded, including the spaces.