Mcp-atlassian

Added
- Added support for Jira timetracking field with proper model field mapping (217, 213)

Changed
- Updated documentation to include fixVersions example in jira_create_issue additional_fields parameter (216)

Fixed
- Changed components parameter type from array to string for Windsurf compatibility (215)

Added
- Enhanced component support for Jira issues to allow multiple components when creating tickets (208, 115)

Fixed
- Fixed variable scope issue in confluence_get_page_children where local variable 'pages' was not accessible due to improper indentation (209, 206)

Added
- Return count metadata (`total`, `start_at`, `max_results`) in Jira search results (`jira_search`, `jira_get_project_issues`, `jira_get_epic_issues`) to enable pagination and count-based queries (203, 164)

Added
- Add pagination support to Jira list tools with new `startAt` parameter for `jira_search`, `jira_get_project_issues`, and `jira_get_epic_issues` (198)
- Add pytest workflow for automated testing across multiple Python versions (3.10, 3.11, 3.12) (200)
- Add Smithery deployment and installation support (194)

Changed
- Unify pagination parameter names to 'startAt' across Jira tools for better consistency and developer experience (201)
- Improve configuration documentation and readability in .env.example with clear sections and alternative options (192)

Fixed
- Fix model property access in JiraIssue to prevent custom fields from overriding model properties (199)
- Fix read_resource signature to properly take in an AnyUrl and return str (195)
- Correct command invocation instructions in README.md to use the `mcp-atlassian` script instead of incorrect `python -m` commands (197)
- Update server tests for read_resource signature change (199)

Added
- Add tools to retrieve Jira Agile boards (`jira_get_agile_boards`), sprints (`jira_get_sprints_from_board`), board issues (`jira_get_board_issues`), and sprint issues (`jira_get_sprint_issues`) (166)
- Implement Jira attachment upload functionality via `jira_update_issue` tool using the `attachments` parameter (168, 7ad3485)
- Add tool to attach content to Confluence pages (`confluence_attach_content`) (178)
- Add explicit `MCPAtlassianAuthenticationError` for handling 401/403 authentication errors, providing clearer feedback for expired tokens or permission issues (190, 183)

Changed
- Improve custom field handling in `JiraIssue` model to automatically include all fields not explicitly processed under `custom_fields`, making more data accessible (182)
- Enable basic authentication support (username/API token) for on-premise/Data Center Confluence instances, previously only token auth was explicitly checked (173)
- General updates and refinements to Jira issue handling logic (187)

Fixed
- Parse `fixVersions` field correctly in Jira issues, previously missed (162)
- Fix JQL error in `list_resources` when user identifier (e.g., email) contains special characters like '' by properly quoting and escaping the identifier (180, 172)
- Fix Confluence search (`confluence_search`) and resource reading (`read_resource` for space URIs) when using space keys that require quotes (e.g., personal spaces '~...', reserved CQL words, numeric keys) by adding conditional CQL identifier quoting (185, 184)
- Fix Python C-level timestamp conversion error ("timestamp too large") when fetching user data from Jira Data Center 9.x instances by using a direct API request bypassing the problematic library parsing (189, 171)

Added
- Added support for retrieving Jira custom fields via MCP API (157)
- Added ability to download attachments from JIRA issues (160)
- Added development environment with devcontainer support (159)
- Added JiraBoards and get_all_agile_boards tool (158)