Registry / llm-agents / mcp-atlassian

mcp-atlassian

JSON →
library0.23.1pypypi✓ verified 22d ago

The Model Context Protocol (MCP) Atlassian integration is an open-source implementation that bridges Atlassian products (Jira and Confluence) with AI language models following Anthropic's MCP specification. This project, currently at version 0.21.1, enables secure, contextual AI interactions with Atlassian tools while maintaining data privacy and security. It is actively developed, with releases focusing on bug fixes and feature enhancements, typically without a strict, public release cadence.

pip install mcp-atlassian
INSTALL
IMPORT
SIG · MCP-ATLASSIAN
M
mcp-atlassian
llm-agentspythonv0.23.1
Install
18.1s avg
Import
Disk
176MB
Pass rate
8/ 10
Env Coverage8 / 10
glibc
3.93.13
musl
3.93.13
Install & Compatibility
Where this runs
tested against v0.23.1 · pip install
no network on importno background threads
Install × environment matrix
Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux
musl
glibc
py 3.10
✓ —
✓ 20.9s
py 3.11
✓ —
✓ 20s
py 3.12
✓ —
✓ 15.7s
py 3.13
✓ —
✓ 15.8s
py 3.9
✕ build_error
✕ build_error
176MB installed
● package 176MB
Code
Verified usage

Verified import paths — ran on the pinned version, not inferred.

AtlassianConfig
from mcp_atlassian.config import AtlassianConfig
AtlassianConfigSchema
from mcp_atlassian.config import AtlassianConfigSchema
AtlassianContextManager
from mcp_atlassian.lib.atlassian_context_manager import AtlassianContextManager

This quickstart demonstrates how to programmatically load the `mcp-atlassian` configuration from environment variables and initialize the `AtlassianContextManager`. It showcases the core configuration and component setup, which is essential for any interaction with the library. Note that actual Atlassian API calls would require valid credentials and specific method calls on the `context_manager` object.

import os from mcp_atlassian.config import AtlassianConfig, AtlassianConfigSchema from mcp_atlassian.lib.atlassian_context_manager import AtlassianContextManager # Set dummy environment variables for demonstration/validation. # In a real scenario, these would be loaded from your shell environment or a .env file. os.environ['ATLASSIAN_SUBDOMAIN'] = os.environ.get('ATLASSIAN_SUBDOMAIN', 'your-subdomain') os.environ['ATLASSIAN_USERNAME'] = os.environ.get('ATLASSIAN_USERNAME', 'email@example.com') os.environ['ATLASSIAN_API_TOKEN'] = os.environ.get('ATLASSIAN_API_TOKEN', 'your-api-token') os.environ['CONFLUENCE_SPACE_ID'] = os.environ.get('CONFLUENCE_SPACE_ID', 'YOURSPACE') os.environ['JIRA_PROJECT_KEY'] = os.environ.get('JIRA_PROJECT_KEY', 'YOURPROJ') try: # Load configuration from environment variables config: AtlassianConfig = AtlassianConfigSchema().load(os.environ) print("Configuration loaded successfully.") print(f"Atlassian Subdomain: {config.atlassian_subdomain}") # Initialize the Atlassian Context Manager # This step demonstrates setup but will require valid credentials for actual API calls. context_manager = AtlassianContextManager(config) print("AtlassianContextManager initialized.") # To perform an actual operation, you would use methods like: # confluence_doc = context_manager.get_confluence_document('page-title') # jira_issue = context_manager.get_jira_issue('JIRA-123') except Exception as e: print(f"Error loading configuration or initializing: {e}") print("Please ensure all required ATLASSIAN_* environment variables are set correctly.")
Debug
Known issues
gotchaConfiguration for `mcp-atlassian` relies heavily on specific environment variables (e.g., `ATLASSIAN_SUBDOMAIN`, `ATLASSIAN_API_TOKEN`, `CONFLUENCE_SPACE_ID`). Incorrectly set or missing variables are the most common source of runtime errors.
fix
Ensure all required environment variables are correctly defined in your shell environment or provided via a `.env` file (which `python-dotenv` can load automatically) before running the application or initializing components.
affects: All versions
gotchaThe library is primarily designed to function as a standalone service or CLI (`mcp-atlassian run-server` or `mcp-atlassian cli`). While individual core components can be imported, their setup and expected operational context often assume the environment is configured as if running the full service, which might not be immediately intuitive for direct programmatic embedding.
fix
For most use cases, it's recommended to interact with the library via its provided CLI or server commands. If programmatic embedding is required, carefully review the official documentation and source code to understand component dependencies and expected environment setup.
affects: All versions
breakingAs a pre-1.0 library (currently `0.21.1`), `mcp-atlassian` does not guarantee strict semantic versioning. Breaking changes to API interfaces, configuration schemas, or core logic may be introduced in minor version increments (e.g., `0.x.y` to `0.x+1.z`).
fix
Pin your `mcp-atlassian` dependency to a specific version (e.g., `mcp-atlassian==0.21.1`) and thoroughly review the release notes or GitHub commit history for any breaking changes when planning an upgrade to a newer minor version.
affects: <1.0.0
Upgrade
Version history
0.23.1latest on PyPI · released Aug 19, 2026
Audit
Dependencies
pydanticrequiredUsed for configuration validation and schema definition.
FlaskrequiredCore web framework for the API server.
connexionrequiredHandles OpenAPI/Swagger specification for the Flask application.
python-dotenvrequiredFacilitates loading environment variables from .env files.
atlassian-python-apirequiredProvides Python bindings for interacting with Atlassian APIs.
Agent activity
24 hits · last 30 days
node
22
OpenAI (training)
1
Resources
mcp-atlassian — pip install mcp-atlassian · libregistry