Version: 0.1.6
Atlassian Agent MCP Server + A2A Agent
Comprehensive AI agent for Jira and Confluence management.
This repository is actively maintained - Contributions are welcome!
The MCP Server can be run in two modes: stdio (for local testing) or http (for networked access).
ATLASSIAN_AGENT_URL: The URL of the target service.ATLASSIAN_AGENT_TOKEN: The API token or access token.
export ATLASSIAN_AGENT_URL="http://localhost:8080"
export ATLASSIAN_AGENT_TOKEN="your_token"
atlassian-mcp --transport "stdio"export ATLASSIAN_AGENT_URL="http://localhost:8080"
export ATLASSIAN_AGENT_TOKEN="your_token"
atlassian-mcp --transport "http" --host "0.0.0.0" --port "8000"export ATLASSIAN_AGENT_URL="http://localhost:8080"
export ATLASSIAN_AGENT_TOKEN="your_token"
atlassian-agent --provider openai --model-id gpt-4o --api-key sk-...docker build -t atlassian-agent .docker run -d \
--name atlassian-agent \
-p 8000:8000 \
-e TRANSPORT=http \
-e ATLASSIAN_AGENT_URL="http://your-service:8080" \
-e ATLASSIAN_AGENT_TOKEN="your_token" \
knucklessg1/atlassian-agent:latestservices:
atlassian-agent:
image: knucklessg1/atlassian-agent:latest
environment:
- HOST=0.0.0.0
- PORT=8000
- TRANSPORT=http
- ATLASSIAN_AGENT_URL=http://your-service:8080
- ATLASSIAN_AGENT_TOKEN=your_token
ports:
- 8000:8000{
"mcpServers": {
"atlassian": {
"command": "uv",
"args": [
"run",
"--with",
"atlassian-agent",
"atlassian-mcp"
],
"env": {
"ATLASSIAN_AGENT_URL": "http://your-service:8080",
"ATLASSIAN_AGENT_TOKEN": "your_token"
}
}
}
}python -m pip install atlassian-agentuv pip install atlassian-agent