fix: make jira_get_sprint_issues respect fields parameter #685
Conversation
The jira_get_sprint_issues MCP tool was ignoring the fields parameter,
returning all custom fields and causing excessive context usage.
Root cause: The method calculated fields_param but never passed it to
the underlying self.jira.get_sprint_issues() call, which doesn't support
a fields parameter in the external atlassian library.
Solution: Replace direct API call with self.search_issues() using JQL
'sprint = {sprint_id}', which properly supports field filtering.
Changes:
- Modified get_sprint_issues to use JQL search instead of direct API call
- Updated existing tests to work with new implementation
- Added test to verify fields parameter works correctly
Benefits:
- Properly respects fields parameter
- Reduces data transfer and token usage
- Consistent with other methods like get_board_issues
- No breaking API changes
|
Hey @jjshanks, thanks for your contribution! Since the original project is no longer actively maintained, we encourage you to move this PR to our active fork. I have a few friends and colleagues helping maintain this fork, and we urge everyone in the community who can't use the Atlassian cloud offering MCP (https://www.atlassian.com/blog/announcements/remote-mcp-server) to help us maintain this version. For our version mcp-atlassian we already have or are about to add:
We have a few more ideas and would love if you can add them to the discussions there. Also would love if you can try it out and help build this with us! Thanks! |
|
This pull request has been automatically marked as stale because it has not had recent activity for 14 days. It will be closed if no further activity occurs. Please leave a comment or remove the 'stale' label if you believe this PR is still relevant. Thank you for your contributions! |
|
Can anyone merge field filtering to the main repo? I'm currently having problem with the context size because of it. And I use it via Docker MCP toolkit, so I'd like to have it fixed in this repository, rather than making fork and fill-in another MCP config. |
|
This pull request has been automatically marked as stale because it has not had recent activity for 14 days. It will be closed if no further activity occurs. Please leave a comment or remove the 'stale' label if you believe this PR is still relevant. Thank you for your contributions! |
3 participants
Description
This PR fixes the
jira_get_sprint_issuesMCP tool to properly respect thefieldsparameter, preventing excessive context usage from returning all customfields regardless of the parameter value.
Root Cause: The method calculated
fields_paramcorrectly but never passed it to the underlyingself.jira.get_sprint_issues()call, which doesn't support afields parameter in the external atlassian library.
Solution: Replace the direct API call with
self.search_issues()using JQLsprint = {sprint_id}, which properly supports field filtering.Fixes: #587
Changes
get_sprint_issuesmethod insrc/mcp_atlassian/jira/search.pyto use JQL search instead of direct API callsearch_issuesinstead ofjira.get_sprint_issues)test_get_sprint_issues_with_fields_parameterto verify fields parameter works correctlyTesting
Built and run locally, verified new results don't eat all the context when using specific fields parameterChecklist