This is an actively maintained community fork of
sooperset/mcp-atlassian.The upstream repository has a growing backlog of unmerged pull requests and unresolved issues. This fork integrates vetted community PRs so users don't have to wait — we monitor upstream issues and PRs and bring them in here when they're ready. All changes are tracked with provenance back to the original PRs and authors.
This is not a takeover. We encourage contributors to submit PRs to upstream first. This fork exists to bridge the gap while the upstream maintainer catches up. When upstream merges a PR, the duplicate here becomes a no-op on the next sync.
Branches:
community/main(default) — upstream + integrated community PRsmain— clean mirror of upstreamteam/main— optional team-specific features (error recovery, extra tools, MCPB extension)Want to help? PRs welcome against
community/main, or submit to upstream — we'll pick it up either way.
Model Context Protocol (MCP) server for Atlassian products (Confluence and Jira). Supports both Cloud and Server/Data Center deployments.
Bitbucket Data Center support is newly added — currently scoped to OAuth 2.0 on Data Center (nine tools today: seven read-only — projects, repositories, and the pull-request review loop (list/get/diff/activities/comments) — plus two pull-request writes, commenting and review status (bitbucket_add_comment, bitbucket_set_review_status, both gated by READ_ONLY_MODE); merge/decline, PAT/basic, and Cloud are future work). See Bitbucket Data Center setup, the Bitbucket tools reference, and the OAuth validation runbook.
mcp-atlassian-jira-demo.mp4
Confluence Demo
confluence-1k.mp4
Go to https://id.atlassian.com/manage-profile/security/api-tokens and create a token.
For Server/Data Center, use a Personal Access Token instead. See Authentication.
Add to your Claude Desktop or Cursor MCP configuration:
{
"mcpServers": {
"mcp-atlassian": {
"command": "uvx",
"args": ["mcp-atlassian"],
"env": {
"JIRA_URL": "https://your-company.atlassian.net",
"JIRA_USERNAME": "your.email@company.com",
"JIRA_API_TOKEN": "your_api_token",
"CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
"CONFLUENCE_USERNAME": "your.email@company.com",
"CONFLUENCE_API_TOKEN": "your_api_token"
}
}
}
}Server/Data Center users: Use
JIRA_PERSONAL_TOKENinstead ofJIRA_USERNAME+JIRA_API_TOKEN. See Authentication for details.
Ask your AI assistant to:
- "Find issues assigned to me in PROJ project"
- "Search Confluence for onboarding docs"
- "Create a bug ticket for the login issue"
- "Update the status of PROJ-123 to Done"
Full documentation is available at mcp-atlassian.soomiles.com.
Documentation is also available in llms.txt format, which LLMs can consume easily:
llms.txt— documentation sitemapllms-full.txt— complete documentation
| Topic | Description |
|---|---|
| Installation | uvx, Docker, pip, from source |
| Authentication | API tokens, PAT, OAuth 2.0 |
| Bitbucket Data Center | OAuth 2.0 setup, tools reference, live-validation runbook |
| Configuration | IDE setup, environment variables |
| HTTP Transport | SSE, streamable-http, multi-user |
| Tools Reference | All Jira & Confluence tools |
| Troubleshooting | Common issues & debugging |
| Product | Deployment | Support |
|---|---|---|
| Confluence | Cloud | Fully supported |
| Confluence | Server/Data Center | Supported (v6.0+) |
| Jira | Cloud | Fully supported |
| Jira | Server/Data Center | Supported (v8.14+) |
| Jira | Confluence |
|---|---|
jira_search - Search with JQL |
confluence_search - Search with CQL |
jira_get_issue - Get issue details |
confluence_get_page - Get page content |
jira_create_issue - Create issues |
confluence_create_page - Create pages |
jira_update_issue - Update issues |
confluence_update_page - Update pages |
jira_transition_issue - Change status |
confluence_add_comment - Add comments |
72 tools total — See Tools Reference for the complete list.
Never share API tokens. Keep .env files secure. See SECURITY.md.
See CONTRIBUTING.md for development setup.
MIT - See LICENSE. Not an official Atlassian product.