Jira MCP Server
Full-featured integration with Jira Cloud, backed by the Jira Cloud REST API (v3) and Agile API (1.0). Provides tools for issue management (get, create, edit, transition, changelogs, watch), comments and worklogs, issue links, JQL search, users/account lookups, projects and issue type/status metadata, project versions and related work, Agile boards and sprints, filters and dashboards, entity properties, and issue attachments (upload, download, delete). Destructive and administrative operations (deleting issues/comments/attachments, creating or updating projects) are opt-in via environment flags.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Jira MCP Server show me all open bugs assigned to me in the current sprint"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Jira MCP Server
A Model Context Protocol (MCP) server for Jira Cloud. It exposes issue, project, board, sprint, search, comment, worklog, link, version, attachment and property operations as MCP tools, backed directly by the Jira Cloud REST API (v3) and Agile API (1.0).
This project is not affiliated with, endorsed by, or sponsored by Atlassian.
Table of Contents
Related MCP server: JIRA MCP Server
Overview
Auth: Jira Cloud API token (email + token, Basic Auth) — no OAuth redirect flow needed.
HTTP client: Node's native
fetch/FormData— no extra HTTP dependency.Runtime deps: just
@modelcontextprotocol/sdkandzod.Destructive/admin tools are off by default:
deleteJiraIssue,deleteJiraComment,deleteJiraIssueAttachment,createJiraProject,updateJiraProjectonly register when you explicitly opt in via env vars (see below).
Prerequisites
Node.js v20 or higher (native
fetch/FormDataneed to be stable).A Jira Cloud site and an API token — create one at https://id.atlassian.com/manage-profile/security/api-tokens.
An MCP-compatible client (VS Code + GitHub Copilot, Claude Code, etc.).
Setup
Clone/copy this project, then install dependencies:
npm install npm run buildOr, once published, install as a dependency of your own project:
npm i @automate-io/jira-mcp-server@latest
## Configuration
The server reads its Jira connection from environment variables — set these in your MCP
client's server config, never commit them to source control.
| Variable | Required | Description |
| --- | --- | --- |
| `JIRA_URL` | Yes | Your site, e.g. `https://your-domain.atlassian.net` |
| `JIRA_USERNAME` | Yes | Account email used to generate the API token |
| `JIRA_TOKEN` | Yes | API token from id.atlassian.com |
| `JIRA_ENABLE_DELETE` | No | Set to `true` to register `deleteJiraIssue`/`deleteJiraComment`/`deleteJiraIssueAttachment` |
| `JIRA_ENABLE_MANAGE` | No | Set to `true` to register `createJiraProject`/`updateJiraProject` |
Example `.vscode/mcp.json` (also included in this repo, using prompted inputs so the token
isn't hardcoded):
```json
{
"servers": {
"jira-mcp": {
"type": "stdio",
"command": "node",
"args": ["node_modules/@automate-io/jira-mcp-server/dist/server.js"],
"cwd": "${workspaceFolder}",
"env": {
"JIRA_URL": "${input:jiraBaseUrl}",
"JIRA_USERNAME": "${input:jiraEmail}",
"JIRA_TOKEN": "${input:jiraApiToken}"
}
}
},
"inputs": [
{ "id": "jiraBaseUrl", "type": "promptString", "description": "Jira Cloud site URL" },
{ "id": "jiraEmail", "type": "promptString", "description": "Jira account email" },
{ "id": "jiraApiToken", "type": "promptString", "description": "Jira API token", "password": true }
]
}Tool catalogue
Grouped by file under src/tools/. 41 tools by default, 46 with delete/manage enabled.
File | Tools |
| getJiraIssue, createJiraIssue, editJiraIssue, listJiraIssueTransitions, transitionJiraIssue, listJiraIssueChangelogs, watchJiraIssue, deleteJiraIssue |
| listJiraIssueComments, addOrEditJiraIssueComment, deleteJiraComment |
| listJiraIssueWorklogs, addOrEditJiraIssueWorklog |
| listJiraIssueLinkTypes, createJiraIssueLink, listJiraIssueRemoteIssueLinks |
| searchJiraIssuesUsingJql |
| getJiraCurrentUser, getJiraUser, lookupJiraAccountId, findJiraIssueAssignableUsers |
| listJiraProjects, listJiraProjectIssueTypesMetadata, getJiraIssueTypeMetaWithFields, listJiraStatuses, listJiraProjectComponents, createJiraProject, updateJiraProject |
| getJiraProjectVersions, manageJiraProjectVersion, getJiraProjectVersionRelatedWork, manageJiraProjectVersionRelatedWork |
| listJiraBoards, getJiraBoardConfig, getJiraBoardIssueData, createJiraBoard |
| listJiraBoardSprints, getJiraBoardSprintData, manageJiraSprint |
| listJiraFilters, listJiraDashboards |
| getJiraEntityProperty, editJiraEntityProperty |
| uploadAttachmentToJiraIssue, downloadJiraIssueAttachment, deleteJiraIssueAttachment |
Italicised tools require the matching JIRA_ENABLE_* flag.
Design notes
src/utilities/jira-client.tsis the single HTTP entry point (jiraRequest). It preserves the real HTTP status code on every response (including errors), parses the body by content-type instead of assuming JSON, and needs no manual context disposal since it's built on nativefetch.src/utilities/tool-helpers.tsprovidessafeHandler(shared try/catch) andtoToolResult(maps a Jira response to MCP content withisErrorreflecting the real status), so each tool handler stays a few lines of Jira-specific logic instead of repeating boilerplate.src/utilities/adf.tsconverts plain text into the minimal Atlassian Document Format Jira Cloud's v3 API requires for rich-text fields (comments, descriptions, worklog comments).searchJiraIssuesUsingJqluses the current token-paginated/rest/api/3/search/jqlendpoint, since the olderstartAt-based/rest/api/3/searchis deprecated.Startup logs go to stderr, never stdout — stdout is reserved for the MCP JSON-RPC stream over the stdio transport, so anything else written there would corrupt it.
Troubleshooting
Server exits immediately with "Missing required environment variable": set
JIRA_URL,JIRA_USERNAME,JIRA_TOKENin your MCP client's server config.401/403 from Jira: regenerate your API token, and confirm
JIRA_USERNAMEmatches the Atlassian account that owns the token.A delete/manage tool doesn't show up: set
JIRA_ENABLE_DELETE/JIRA_ENABLE_MANAGEto"true"in the server'senvconfig.
Available Tools
41 toolsaddOrEditJiraIssueCommentC
Add a comment, or edit an existing one.
Inputs: body (required), issueIdOrKey (required), commentId, visibilityType, +1 more.
addOrEditJiraIssueWorklogC
Log time on an work item, or edit an existing worklog.
Inputs: issueIdOrKey (required), comment, started, reduceBy, +5 more.
createJiraBoardB
Create a company-managed board (scrum or kanban) from an existing filter.
Inputs: name (required), type (required), filterId (required), projectKeyOrId.
createJiraIssueC
Create a new Jira work item.
Inputs: summary (required), projectKey (required), fields, update, +3 more.
createJiraIssueLinkC
Create a link between two Jira work items.
Inputs: linkTypeName (required), inwardIssueKey (required), outwardIssueKey (required), comment.
downloadJiraIssueAttachmentBRead-onlyIdempotent
Get a short-lived download URL for an attachment and a local command to save the file.
Inputs: attachmentId (required), savePath.
editJiraEntityPropertyBIdempotent
Set a property in the key/value store on an work item, comment, worklog, dashboard item, or user.
Inputs: entityType (required), propertyKey (required), value, entityId, +3 more.
editJiraIssueCIdempotent
Edit an existing work item; only fields editable on that work item can be set.
Inputs: issueIdOrKey (required), fields, update, notifyUsers.
findJiraIssueAssignableUsersCRead-onlyIdempotent
Find users assignable to a Jira space or work item.
Inputs: query, maxResults, projectKey, issueIdOrKey.
getJiraBoardConfigBRead-onlyIdempotent
A board's configuration: filter, column layout, and each column's mapped statuses.
Inputs: boardId (required).
getJiraBoardIssueDataCRead-onlyIdempotent
The work items on a board, its backlog, or both, with optional count and JQL filtering.
Inputs: boardId (required), jql, fields, source, +2 more.
getJiraBoardSprintDataBRead-onlyIdempotent
Resolve a board plus its sprint data in one call; defaults to the active sprint.
Inputs: boardId (required), sprintId, includeIssues.
getJiraCurrentUserBRead-onlyIdempotent
Details for the current Jira user.
Inputs: no named parameters.
getJiraEntityPropertyBRead-onlyIdempotent
Read a property from the key/value store on an work item, comment, worklog, dashboard item, or user.
Inputs: entityType (required), propertyKey (required), entityId, accountId, +2 more.
getJiraIssueCRead-onlyIdempotent
Get a Jira work item by ID or key.
Inputs: issueIdOrKey (required), expand, fields, properties.
getJiraIssueTypeMetaWithFieldsBRead-onlyIdempotent
Schema and field metadata for a specific work item type in a project, for creating work items with correct fields.
Inputs: issueTypeId (required), projectIdOrKey (required).
getJiraProjectVersionRelatedWorkCRead-onlyIdempotent
Related work links (evidence) on a space version (release).
Inputs: versionId (required).
getJiraProjectVersionsCRead-onlyIdempotent
A space's releases/versions, or a single version.
Inputs: startAt, versionId, maxResults, projectIdOrKey.
getJiraUserBRead-onlyIdempotent
Retrieves a Jira user's profile by account ID.
Inputs: accountId (required).
listJiraBoardsCRead-onlyIdempotent
Lists boards visible to the current user.
Inputs: name, type, startAt, maxResults, +1 more.
listJiraBoardSprintsBRead-onlyIdempotent
Sprints from a board, optionally filtered to active, future, or closed.
Inputs: boardId (required), state, startAt, maxResults.
listJiraDashboardsCRead-onlyIdempotent
Dashboards visible to the current user, for discovery and linking workflows.
Inputs: startAt, maxResults, dashboardName.
listJiraFiltersBRead-onlyIdempotent
Search saved filters visible to the current user, returning IDs and JQL.
Inputs: startAt, accountId, filterName, maxResults.
listJiraIssueChangelogsBRead-onlyIdempotent
An work item's field-change history, paginated oldest-first.
Inputs: issueIdOrKey (required), startAt, maxResults.
listJiraIssueCommentsCRead-onlyIdempotent
Paginated comments for a Jira work item.
Inputs: issueIdOrKey (required), expand, orderBy, startAt, +1 more.
listJiraIssueLinkTypesCRead-onlyIdempotent
Available work item link types in Jira.
Inputs: no named parameters.
listJiraIssueRemoteIssueLinksCRead-onlyIdempotent
Remote links associated with a Jira work item.
Inputs: issueIdOrKey (required).
listJiraIssueTransitionsCRead-onlyIdempotent
Available workflow transitions for a Jira work item.
Inputs: issueIdOrKey (required).
listJiraIssueWorklogsBRead-onlyIdempotent
Paginated worklogs (logged time) for a Jira work item.
Inputs: issueIdOrKey (required), startAt, maxResults, startedAfter, +1 more.
listJiraProjectComponentsCRead-onlyIdempotent
Components in a space, with IDs, names, work item counts, and ownership details.
Inputs: projectIdOrKey (required).
listJiraProjectIssueTypesMetadataCRead-onlyIdempotent
All work item types available in a Jira space.
Inputs: projectIdOrKey (required).
listJiraProjectsCRead-onlyIdempotent
Get Jira spaces visible to the current user.
Inputs: query, startAt, maxResults.
listJiraStatusesBRead-onlyIdempotent
List of statuses a Jira work item can be in.
Inputs: projectIdOrKey.
lookupJiraAccountIdBRead-onlyIdempotent
Look up a Jira user account ID by display name or email.
Inputs: query (required), maxResults.
manageJiraProjectVersionC
Create, update, release, or archive a space version (release or fix version).
Inputs: operation (required), name, startDate, versionId, +3 more.
manageJiraProjectVersionRelatedWorkC
Create or update related work links on a space version.
Inputs: url (required), title (required), category (required), versionId (required), +1 more.
manageJiraSprintC
Create, populate, start, update, or close a sprint.
Inputs: operation (required), goal, name, endDate, +4 more.
searchJiraIssuesUsingJqlCRead-onlyIdempotent
Search Jira work items using JQL.
Inputs: jql (required), expand, fields, maxResults, +1 more.
transitionJiraIssueB
Transition an work item to a new status, and/or assign it to a sprint or backlog.
Inputs: issueIdOrKey (required), transitionId (required), fields, comment.
uploadAttachmentToJiraIssueC
Attach a local file to an work item.
Inputs: filePath (required), issueIdOrKey (required), mimeType.
watchJiraIssueCIdempotent
Watch or unwatch a work item.
Inputs: watch (required), issueIdOrKey (required), accountId.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 41 tools
Most tools have clearly distinct purposes, but some overlap exists among board/sprint/issue-data retrieval tools (e.g., getJiraBoardIssueData vs searchJiraIssuesUsingJql) and version tools. Descriptions generally clarify boundaries well enough for an agent to select correctly.
Names mostly follow a verb + Jira + noun pattern (e.g., getJiraIssue, listJiraProjects, createJiraIssue). Minor deviations like uploadAttachmentToJiraIssue and searchJiraIssuesUsingJql break the pattern slightly but remain readable and predictable.
41 tools is heavy for a single MCP server and exceeds the 25+ threshold for being too many. Many operations could be consolidated (e.g., list/get/manage variants across issues, boards, and versions).
The surface covers broad read/write operations for issues, comments, worklogs, links, attachments, users, projects, versions, boards, sprints, filters, dashboards, and properties. However, delete operations are largely missing for issues, comments, worklogs, links, attachments, and boards/sprints/filters/dashboards, leaving notable lifecycle gaps.
Related MCP Connectors
Read and write Mission Control state via MCP — projects, tasks, subtasks, templates, status updates.
Unofficial Confluence MCP — wraps the Confluence Cloud REST API v2 (OAuth)
One OAuth connector for your team's remote MCP servers, with Tool Packs, guardrails and logs.
Unofficial community ClickUp MCP via personal API token. Not affiliated with ClickUp.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceMCP server for interacting with Jira Cloud instances. Enables issue management, JQL queries, project and sprint management, and batch operations via natural language interfaces.141 npm4MIT
- AlicenseNot gradedqualityDmaintenanceEnables natural language interaction with JIRA through MCP, providing 35 tools for issues, comments, transitions, projects, boards, sprints, epics, links, worklogs, versions, attachments, users, and fields.MIT
- AlicenseNot gradedqualityDmaintenanceManages Atlassian Jira Cloud projects, issues, sprints, boards, worklogs, comments, and workflow transitions from MCP-compatible clients.MIT
- AlicenseNot gradedqualityDmaintenanceJira Cloud MCP server providing Jira-first tools for common workflows and full REST API coverage through a generic request tool.MIT