Atlassian MCP Detailed Setup
Advanced configuration and detailed documentation for the Atlassian MCP server.
For basic setup, see MCP Setup - Atlassian MCP.
Separate Confluence and Jira Configurations
The quick start configures both services together. You can also configure them separately:
Using Shell Environment Variables
Claude Code supports `${VAR}` syntax for environment variable interpolation at runtime. This approach ensures tokens are never persisted in configuration files.
-
Set environment variables in your shell profile (
~/.bashrc,~/.zshrc, etc.):export CONFLUENCE_PERSONAL_TOKEN="your-confluence-token-here" export JIRA_PERSONAL_TOKEN="your-jira-token-here" -
Configure MCP without embedding tokens:
claude mcp add maven-atlassian -s user \ -e 'CONFLUENCE_URL=https://cwiki.apache.org/confluence' \ -e 'CONFLUENCE_PERSONAL_TOKEN=${CONFLUENCE_PERSONAL_TOKEN}' \ -e 'JIRA_URL=https://issues.apache.org/jira' \ -e 'JIRA_PERSONAL_TOKEN=${JIRA_PERSONAL_TOKEN}' \ -- docker run -i --rm \ -e CONFLUENCE_URL \ -e CONFLUENCE_PERSONAL_TOKEN \ -e JIRA_URL \ -e JIRA_PERSONAL_TOKEN \ ghcr.io/sooperset/mcp-atlassian:latest
The `${VAR}` references are expanded at runtime from your shell environment.
|
If an environment variable is not set when Claude Code starts, configuration parsing will fail. Ensure your tokens are exported before launching Claude Code. |
Security Best Practices
|
Never commit Personal Access Tokens to version control. Always use the `${VAR}` interpolation syntax to keep tokens out of configuration files. |