Pydantic ValidationError in mcp-atlassian jira_search when descriptions are in ADF format
Using mcp-atlassian 0.11.10 on Windows 11 (Python 3.13) from Cline 1.1.10 in IntelliJ IDEA,
jira_searchagainst Jira Cloud fails withValidationError: 1 validation error for JiraIssue / description / Input should be a valid string [type=string_type, input_value={'type': 'doc', 'version': 1, ...}, input_type=dict]when issues have ADF-formatted descriptions. How can I query these issues without the Pydantic error?Originally asked on GitHub: https://github.com/sooperset/mcp-atlassian/issues/785
3 answers
Fixed in mcp-atlassian v0.11.11; the maintainer closed this issue as completed and pointed to PR #782. The reporter traced the failure to the
JiraIssuemodel'sdescriptionfield accepting only a plain string while Jira Cloud returns descriptions as ADF dicts ({"type": "doc", "version": 1, ...}). Per the maintainer, the fix adds an ADF parser that properly handles the node types that caused the Pydantic validation errors.Upgrade to the fixed release:
bashpip install --upgrade mcp-atlassianThe reporter also suggested a stopgap for versions below v0.11.11: exclude
descriptionfrom thefieldsargument so the ADF payload is never requested:pythonjira_search( jql="assignee = currentUser() AND status != Done", fields="summary,status,assignee,priority,created,updated,issuetype", limit=10 )The thread does not state that this workaround was actually tested, so treat it as a suggested mitigation, not a verified fix. Note that the reported failure occurs specifically when the returned issues have ADF-formatted descriptions, not for every search that includes the
descriptionfield.Did this work for you?What a GitHub user tried on GitHub (Outcome unclear)
The reporter suggested omitting
descriptionfromfields(e.g.fields="summary,status,assignee,priority,created,updated,issuetype") so the ADF payload is never requested. The thread does not say whether this was tested.Source: https://github.com/sooperset/mcp-atlassian/issues/785 (opens external site)
Did this work for you?What @sooperset tried on GitHub (Outcome unclear)
The maintainer announced the fix in v0.11.11 (PR #782) and advised
pip install --upgrade mcp-atlassian, explaining that the ADF parser now handles the node types that caused the validation errors. No user in the thread reports actually running the upgrade.Source: https://github.com/sooperset/mcp-atlassian/issues/785#issuecomment-3698336072 (opens external site)
Did this work for you?
Related questions
- mcp-atlassian: 'JiraConfig' object has no attribute 'read_only' when calling create_issue
- mcp-atlassian Docker fails on startup with 'ValueError: I/O operation on closed file' — fixed?
- OAuth 2.0 for Confluence Data Center (on-prem) in mcp-atlassian?
- mcp-atlassian crashes with Pydantic 2.12.0: 'cannot specify both default and default_factory'
- mcp-atlassian Jira search_issues: 'requested API has been removed' error and total -1