MCP SERVERDrive Jira from an assistant: file and re-order tickets, hang them off epics, walk a sprint through its states, and read the release and component progress that follows.
LABSCO SUMMARY
With the instance mapping in place, an agent can file the ticket, park it under an epic, drop it into the running sprint, and then answer how that sprint is tracking. The instance and working_dir arguments are not optional in practice, so the mapping file is where a setup succeeds or fails. Roadmap-level plan tools depend on Jira Premium, and confluence_health_check answers for a different server altogether.
A multi-instance Jira client whose tool set covers issues, agile boards, epics, versions, components, JQL and Premium strategic planning.
A Jira instance configured in .jira-config.json with its project mapping, since the instance name and a working_dir path travel with every call; detect_project_fields resolves the custom field IDs that file needs. Plans require Jira Premium, and removing an issue outright requires the Delete Issues permission.
Build from source — git clone https://github.com/cfdude/mcp-jira.git
USE CASES
Create the issue with its summary, type and priority, attach an estimate, and fold in whatever context surfaces afterwards.
Pull the project's issues, re-rank them against reference issues, apply shared field changes in one request, and push the top of the list into the running sprint.
Confirm where the sprint stands, close it, read the report it leaves behind, and create its successor on the same board.
Component progress shows where work sits by feature area; the version record and its breakdown show what the release itself still owes.
Create the plan, assign the teams that own the work, and check what is on it already. These calls depend on Jira Premium.
Search for the project, read what it contains, then list its boards and inspect the columns of the one that looks right.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
add_commentAdd a comment to an existing Jira issue. CRITICAL FOR AI: ALWAYS specify 'instance' parameter and use exact issue key from list_issues. Example: {working_dir: '/path', instance: 'onvex', issue_key: 'JOB-123', comment: 'Updated requirements'}
add_plan_teamAdd a team to a strategic plan (Jira Premium feature)
archive_planArchive a strategic plan (Jira Premium feature)
bulk_update_issuesUpdate multiple issues at once. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact issue keys from list_issues. Example: {working_dir: '/path', instance: 'onvex', issueKeys: ['JOB-1', 'JOB-2'], updates: {status: 'In Progress'}}
complete_sprintComplete/close an active sprint. CRITICAL FOR AI: ALWAYS specify 'instance' parameter. Use get_sprint_details first to verify sprint is active. Example: {working_dir: '/path', instance: 'onvex', sprintId: 335}
confluence_health_checkCheck the health and connectivity status with the Confluence MCP server from Jira's perspective. Returns connection information and integration capabilities.
create_componentCreate a new project component for organizing work by feature areas. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB', name: 'Frontend UI'}
create_epicCreate a new Epic issue. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Epics organize related stories/tasks. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB', name: 'User Auth', summary: 'Authentication System Epic'}
create_epic_with_issuesCreate an epic with linked issues in a single operation. Reduces API calls and ensures proper linking. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB', epic: {name: 'User Auth', summary: 'Authentication System'}, issues: [{summary: 'Login API', type: 'Task'}, {summary: 'User Registration', type: 'Story'}]}
create_filterCreate saved filters for consistent project tracking and reporting. CRITICAL FOR AI: ALWAYS specify 'instance'. Test JQL first with search_issues_jql. Example: {working_dir: '/path', instance: 'onvex', name: 'Sprint Planning', jql: 'project = JOB'}
create_issueCreate a new Jira issue. IMPORTANT FOR AI: Always specify 'instance' and 'projectKey' parameters explicitly to avoid configuration errors. Example: {working_dir: '/path/to/project', instance: 'onvex', projectKey: 'JOB', summary: 'Fix login bug', type: 'Task'}
create_planCreate a new strategic plan for high-level roadmap management (Jira Premium feature)
create_sprintCreate a new sprint in a Jira board. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Use list_boards first to verify board exists. Creates sprint in 'future' state - use update_sprint to start it. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB', name: 'Sprint 1', goal: 'MVP features'}
create_versionCreate a new project version/release for organizing work by milestones. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB', name: 'v1.0.0'}
delete_issueDelete a Jira issue. WARNING: Requires 'Delete Issues' permission. Only issue creators and project administrators can delete issues. Consider using status changes or labels instead of deletion.
detect_project_fieldsDetect custom field IDs (story points, sprint, epic link) for a specific project and instance. Provides ready-to-copy configuration snippets for .jira-config.json setup. CRITICAL FOR AI: ALWAYS specify 'instance' parameter for multi-instance support. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
duplicate_planDuplicate a strategic plan to create a new plan based on existing configuration (Jira Premium feature)
estimate_issueSet estimation value for an issue. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact issue key. Example: {working_dir: '/path', instance: 'onvex', issueKey: 'JOB-1', value: '5'}
get_board_configurationGet detailed configuration of a board including columns and settings. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102}
get_board_cumulative_flowGet cumulative flow diagram data for a board. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102}
get_board_reportsGet board reports and current metrics. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102}
get_burndown_chart_dataGet burndown chart data for a specific sprint. CRITICAL FOR AI: ALWAYS specify 'instance' and use sprintId from create_sprint. Example: {working_dir: '/path', instance: 'onvex', sprintId: 335}
get_component_progressGet detailed component progress with issue distribution and completion metrics. CRITICAL FOR AI: ALWAYS specify 'instance' and use componentId from list_components. Example: {working_dir: '/path', instance: 'onvex', componentId: '10456'}
get_issueGet detailed information about a specific Jira issue. CRITICAL FOR AI: ALWAYS specify 'instance' parameter. Use this to get current status, description, and fields before updating. Example: {working_dir: '/path', instance: 'onvex', issue_key: 'JOB-123'}
get_issue_typesGet available issue types, their hierarchy, and required fields. CRITICAL FOR AI: ALWAYS specify 'instance' and usually 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
get_planGet detailed information about a specific strategic plan (Jira Premium feature)
get_plan_teamsGet detailed information about teams assigned to a strategic plan (Jira Premium feature)
get_project_detailsGet comprehensive project information including components, versions, and features. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
get_project_statusesGet comprehensive project workflow statuses and transitions. CRITICAL FOR AI: ALWAYS specify 'instance' and usually 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
get_sprint_detailsGet detailed information about a sprint including progress and issues. CRITICAL FOR AI: ALWAYS specify 'instance' parameter and use sprintId from create_sprint or board reports. Example: {working_dir: '/path', instance: 'onvex', sprintId: 335}
get_sprint_reportGet comprehensive sprint report with metrics and analysis. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102, sprintId: 335}
get_transitionsGet available workflow transitions for a Jira issue. CRITICAL FOR AI: Use this BEFORE attempting to change issue status. ALWAYS specify 'instance' parameter. Example: {working_dir: '/path', instance: 'onvex', issueKey: 'JOB-123'}
get_velocity_chart_dataGet velocity chart data for team performance analysis. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102}
get_version_progressGet detailed version progress including issue counts, status breakdown, and timeline analysis. CRITICAL FOR AI: ALWAYS specify 'instance' and use versionId from list_versions. Example: {working_dir: '/path', instance: 'onvex', versionId: '10123'}
jira_health_checkGet comprehensive health information for this Jira MCP server including uptime, cross-server integration status, and server capabilities. Essential for monitoring and troubleshooting cross-server communication.
list_boardsList all Jira boards for a project. CRITICAL FOR AI: ALWAYS specify 'instance' and usually 'projectKey' to filter results. Use this first to find board IDs for sprint operations. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
list_componentsList project components for feature-based work organization. CRITICAL FOR AI: ALWAYS specify 'instance' and usually 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
list_custom_fieldsList all available custom and system fields for a Jira instance. Shows field IDs, types, and provides configuration examples. Useful for discovering what fields can be added to your configuration.
list_epic_issuesList all issues in an epic. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact epicKey from create_epic. Example: {working_dir: '/path', instance: 'onvex', epicKey: 'JOB-1'}
list_instancesList all configured Jira instances and their project mappings. CRITICAL FOR AI: Use this FIRST to discover available instances before any other operations. Shows which projects are mapped to which instances. No additional parameters needed besides working_dir. Example: {working_dir: '/path/to/project'}
list_issuesList issues in a specific Jira project. CRITICAL FOR AI: ALWAYS specify 'instance' and 'projectKey' parameters. Use this to see existing issues before creating new ones or to get issue keys for updates. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
list_plansList strategic plans for high-level roadmap and portfolio management (Jira Premium feature). CRITICAL FOR AI: ALWAYS specify 'instance'. Example: {working_dir: '/path', instance: 'onvex'}
list_versionsList project versions for release planning and milestone tracking. CRITICAL FOR AI: ALWAYS specify 'instance' and usually 'projectKey'. Example: {working_dir: '/path', instance: 'onvex', projectKey: 'JOB'}
manage_board_quickfiltersList or get details of board quickfilters. CRITICAL FOR AI: ALWAYS specify 'instance' and use boardId from list_boards. Example: {working_dir: '/path', instance: 'onvex', boardId: 102, action: 'list'}
move_issues_to_epicMove issues to an epic. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact keys from list_issues and create_epic. Example: {working_dir: '/path', instance: 'onvex', epicKey: 'JOB-1', issueKeys: ['JOB-5', 'JOB-6']}
move_issues_to_sprintAdd issues to a sprint. CRITICAL FOR AI: ALWAYS specify 'instance' parameter. Use list_issues first to get issue keys, and ensure sprint exists. Example: {working_dir: '/path', instance: 'onvex', sprintId: 123, issueKeys: ['JOB-1', 'JOB-2']}
rank_epicsRank epics relative to each other. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact epic keys. Example: {working_dir: '/path', instance: 'onvex', epicToRank: 'JOB-1', rankAfterEpic: 'JOB-2'}
rank_issuesRank multiple issues relative to other issues. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact issue keys. Example: {working_dir: '/path', instance: 'onvex', issues: ['JOB-1'], rankAfterIssue: 'JOB-2'}
remove_plan_teamRemove a team from a strategic plan (Jira Premium feature)
search_issues_jqlAdvanced issue search using Jira Query Language (JQL) for powerful project analysis. CRITICAL FOR AI: ALWAYS specify 'instance'. Example: {working_dir: '/path', instance: 'onvex', jql: 'project = JOB AND status != Done'}
search_projectsSearch and discover projects across the Jira instance with advanced filtering. CRITICAL FOR AI: ALWAYS specify 'instance'. Example: {working_dir: '/path', instance: 'onvex', query: 'mobile'}
transition_issuePerform a workflow transition on a Jira issue. CRITICAL FOR AI: ALWAYS use get_transitions first to get valid transition IDs. ALWAYS specify 'instance' parameter. Example: {working_dir: '/path', instance: 'onvex', issueKey: 'JOB-123', transitionId: '5'}
trash_planMove a strategic plan to trash (soft delete) (Jira Premium feature)
update_epic_detailsUpdate epic details using Agile API. CRITICAL FOR AI: ALWAYS specify 'instance' and use exact epicKey from list_epic_issues. Example: {working_dir: '/path', instance: 'onvex', epicKey: 'JOB-1', name: 'Updated Epic Name'}
update_issueUpdate an existing Jira issue. CRITICAL FOR AI: ALWAYS specify 'instance' parameter and use exact issue key from list_issues. Commonly used to change status, assignee, or add comments. Example: {working_dir: '/path', instance: 'onvex', issue_key: 'JOB-123', status: 'In Progress'}
update_planUpdate an existing strategic plan using JSON Patch operations (Jira Premium feature)
update_sprintUpdate sprint details or change sprint state (future→active→closed). CRITICAL FOR AI: ALWAYS specify 'instance' parameter. To activate a sprint, MUST provide both startDate and endDate. Example: {working_dir: '/path', instance: 'onvex', sprintId: 123, state: 'active', startDate: '2025-07-29', endDate: '2025-08-12'}
BEFORE YOU INSTALL
57 tools when we connected. The documentation covers 42; the 15 undocumented ones are mostly Jira Plans, epic and board tools. Plan tools register but need Jira Premium to work.
All 3 documented clients run the server on your machine.
CONTEXT COST
Rated by the size of the 57 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
You clone and build this one yourself.
git clone https://github.com/cfdude/mcp-jira.git
cd mcp-jira
npm install
npm run buildCreate .jira-config.json in that folder — the client configurations below start the server there — with your Jira instance (email, apiToken, domain) and the project keys it serves, in the format of the README's Multi-Instance Configuration; detect_project_fields reports the custom field IDs to add.
{
"mcpServers": {
"jira": {
"command": "node",
"args": [
"<YOUR_PATH_TO_JIRA_SERVER>/build/index.js"
],
"cwd": "<YOUR_PATH_TO_JIRA_SERVER>",
"env": {
"JIRA_EMAIL": "<YOUR_JIRA_EMAIL>",
"JIRA_API_TOKEN": "<YOUR_JIRA_API_TOKEN>",
"JIRA_DOMAIN": "<YOUR_JIRA_DOMAIN>"
},
"disabled": false,
"alwaysAllow": true
}
}
}<YOUR_PATH_TO_JIRA_SERVER> (in args and cwd) is the folder you cloned and built.
{
"mcpServers": {
"jira": {
"command": "node",
"args": [
"<YOUR_PATH_TO_JIRA_SERVER>/build/index.js"
],
"cwd": "<YOUR_PATH_TO_JIRA_SERVER>",
"env": {
"JIRA_EMAIL": "<YOUR_JIRA_EMAIL>",
"JIRA_API_TOKEN": "<YOUR_JIRA_API_TOKEN>",
"JIRA_DOMAIN": "<YOUR_JIRA_DOMAIN>"
},
"disabled": false,
"alwaysAllow": [
"create_issue",
"list_issues",
"update_issue",
"get_issue",
"delete_issue",
"add_comment",
"list_instances",
"create_component",
"list_components",
"get_component_progress",
"create_version",
"list_versions",
"get_version_progress",
"search_issues_jql",
"search_projects",
"create_filter",
"get_project_details",
"get_project_statuses",
"get_issue_types",
"list_plans",
"create_sprint",
"update_sprint",
"get_sprint_details",
"create_epic",
"update_epic_details",
"bulk_update_issues"
]
}
}
}<YOUR_PATH_TO_JIRA_SERVER> (in args and cwd) is the folder you cloned and built.
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"jira": {
"type": "local",
"command": [
"node",
"build/index.js"
],
"enabled": true,
"environment": {
"JIRA_CONFIG_PATH": "./.jira-config.json"
}
}
}
}build/index.js and ./.jira-config.json here are relative: keep this opencode.json in your clone, next to the .jira-config.json from the step above, or change both to absolute paths. JIRA_CONFIG_PATH can be absolute, relative to this config file, or start with ~.
MCP servers
InfinitIQ-Tech
guanghuang
Dsazz
aashari