anthonyd23/atlassian_mcp
If you are the rightful owner of atlassian_mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.
The Atlassian MCP Server is a Model Context Protocol server designed for integration with Atlassian tools such as Jira, Confluence, and Bitbucket. It supports both Atlassian Cloud and Data Center/Server deployments, offering a comprehensive suite of 46 tools to enhance productivity and streamline workflows.
Atlassian MCP Server
Model Context Protocol (MCP) server for Atlassian tools (Jira, Confluence, and Bitbucket).
Enterprise-grade MCP server providing 101 production-ready tools for Jira, Confluence, and Bitbucket
- ✨ Works with Amazon Q Developer, Claude, Cursor, and more
- 🚀 Deploy locally or to AWS Lambda
- 🔒 Enterprise security built-in
- 🎫 Ticket Support Agent with 6 specialized tools
Quick Links
- 🚀 Quick Start - Get running in 5 minutes
- 📖 - Learn how MCP works
- 🤖 - Connect to Amazon Q, Claude, Cursor, etc.
- ☁️ - Deploy to Lambda
- 📊 - CloudWatch metrics and alerts
- 🏗️ - System design and components
Quick Start
Local Development:
pip install -r mcp_server/requirements.txt
cp config.template.yaml config.yaml
# Edit config.yaml with your credentials
python mcp_server/main.py
Configuration Examples:
Cloud:
deployment_type: cloud
cloud:
atlassian_base_url: https://yourcompany.atlassian.net
atlassian_username: your-email@company.com
atlassian_api_token: your-token
bitbucket_workspace: your-workspace
bitbucket_api_token: your-token
Data Center:
deployment_type: datacenter
datacenter:
jira_base_url: https://jira.company.com
jira_pat_token: your-token
confluence_base_url: https://wiki.company.com
confluence_pat_token: your-token
bitbucket_base_url: https://git.company.com
bitbucket_pat_token: your-token
bitbucket_project: PROJECT_KEY
AWS Deployment:
cp config.template.yaml config.yaml
# Edit config.yaml with your credentials
python deploy.py
Features
- Jira (31 tools): Issues, comments, transitions, attachments, attachment upload, users, worklogs, labels, issue linking, advanced search, priority management, agile boards, sprints, user permissions
- Confluence (30 tools): Pages, spaces, comments, attachments, search, users, labels, page history, permissions, page copying, user content, recent content, version restore, search by author/label, page hierarchy (move, children, descendants, ancestors), CQL search
- Bitbucket (34 tools): Repositories, pull requests, commits, branches, diffs, file search, reviewers, branch management, PR activity, default reviewers, author filtering, change requests, branch restrictions, build status, webhooks
- Ticket Support Agent (6 tools): Open ticket triage, template validation, assignee suggestions, team workload analysis, expertise JQL construction, troubleshooting doc lookup
- Flexible Credentials: Configure only the services you need
- Dual Platform: Supports both Cloud and Data Center deployments
- AWS Ready: Deploy as Lambda function with API Gateway
Prerequisites
Local Development:
- Python 3.11+
- Atlassian account (Cloud or Data Center)
- API tokens for services you want to use
AWS Deployment (optional):
- AWS CLI configured with credentials
- AWS SAM CLI installed (installation guide)
Generate Tokens:
- Cloud: https://id.atlassian.com/manage-profile/security/api-tokens
- Bitbucket Cloud: https://bitbucket.org/account/settings/app-passwords/
- Data Center: Profile → Personal Access Tokens
Configuration
AWS Deployment
- Copy the template:
cp config.template.yaml config.yaml
- Edit
config.yamlwith your credentials:
For Cloud:
deployment_type: cloud
cloud:
atlassian_base_url: https://yourcompany.atlassian.net
atlassian_username: your-email@company.com
atlassian_api_token: your-token
bitbucket_workspace: your-workspace # optional
bitbucket_api_token: your-token # optional
For Data Center:
deployment_type: datacenter
datacenter:
jira_base_url: https://jira.company.com
jira_pat_token: your-token
confluence_base_url: https://wiki.company.com
confluence_pat_token: your-token
bitbucket_base_url: https://git.company.com
bitbucket_pat_token: your-token
bitbucket_project: PROJECT_KEY
- Deploy:
python deploy.py
Local Development
For local development, use config.yaml (same format as AWS deployment):
cp config.template.yaml config.yaml
# Edit config.yaml with your credentials and optional ticket support agent config
python mcp_server/main.py
Note: The server loads configuration from config.yaml automatically. Environment variables can be used as an alternative if config.yaml is not present, but using config.yaml is the recommended approach for consistency with AWS deployment.
Platform Detection: The server automatically detects whether to use Cloud or Data Center APIs in this order:
DEPLOYMENT_TYPEenvironment variable (cloudordatacenter)deployment_typefield inconfig.yaml- Presence of Data Center credentials (PAT tokens)
- Defaults to Cloud if none of the above
AI Agent Integration
Integrate with popular AI agents and development tools:
Amazon Q Developer:
{
"mcpServers": {
"atlassian-mcp": {
"command": "python",
"args": ["/absolute/path/to/mcp_server/main.py"],
"env": {
"ATLASSIAN_BASE_URL": "https://yourcompany.atlassian.net",
"ATLASSIAN_USERNAME": "your-email@company.com",
"ATLASSIAN_API_TOKEN": "your-token"
}
}
}
}
Also supports: Claude Desktop, Cline (VS Code), Cursor, Continue, Zed Editor
See for complete setup instructions.
AWS Deployment
- Configure credentials:
cp config.template.yaml config.yaml
# Edit config.yaml with your Atlassian credentials
- Deploy:
python deploy.py
The script will:
- Build the SAM application
- Deploy to AWS with your credentials
- Display the MCP API URL
See for detailed instructions.
Testing
Unit Tests:
pip install -r requirements-dev.txt
pytest tests/unit/
Integration Tests:
Cloud (tests with real credentials):
python tests/cloud/test_all_cloud_tools.py
Data Center (tests with real credentials):
python tests/datacenter/test_all_dc_tools.py
Common/Agent tools (tests with real credentials):
python tests/common/test_all_common_tools.py
Test Features:
- Comprehensive unit test coverage
- Integration tests for core workflows
- Verbose output showing results
Monitoring
CloudWatch integration with:
- Structured JSON logging
- Custom metrics (tool usage, duration)
- Automatic alarms (errors, throttles, slow responses)
- Dashboard for visualization
See for setup and configuration.
Documentation
- - Understanding MCP and how this server works
- - AI agent integration (Claude, Cursor, Cline, etc.)
- - AWS deployment and configuration
- - Comprehensive testing guide
- - Ticket support agent tools
- - CloudWatch setup and alerts
- - System architecture and design
API Documentation
- Jira Cloud
- Confluence Cloud
- Bitbucket Cloud
- Jira Data Center
- Confluence Data Center
- Bitbucket Data Center
Security
- IAM authentication for same-account access
- Rate limiting (100 req/sec, 200 burst)
- Encrypted credentials in Lambda environment variables
- HTTPS-only traffic
- config.yaml gitignored (credentials not in version control)
License
This project is licensed under the MIT License - see the file for details.
Acknowledgments
- Built on the Model Context Protocol specification
- Integrates with Atlassian Cloud and Data Center APIs
- Designed for Amazon Q Developer and other MCP-compatible AI assistants