Jira MCP Server

An Apify Actor that exposes Jira Cloud as a set of MCP-style tools for AI assistants. Each run executes one or more tool calls against your Jira Cloud instance and writes the results to the Actor's dataset. It covers 14 tools spanning issue search, creation, updates, transitions, comments, assignment, sprint management, and project listing.

Data source: Jira Cloud REST API v3 (/rest/api/3/...) plus the Jira Agile API 1.0 (/rest/agile/1.0/...) for board and sprint tools. Authentication is Atlassian Basic auth using your account email and an API token.

Features

• JQL search — run Jira Query Language searches (jira_search)
• Issue read/create/update — fetch, create, and update issues (no delete)
• Status transitions — move issues through workflow states by transition name
• Comments — add and read comments (plain text)
• Assignment — assign or unassign issues by Atlassian account ID
• Sprint tools — list a board's sprints and list issues in a sprint (requires Jira Software)
• Project listing — list accessible projects
• My issues — list issues assigned to the authenticated user

Input

Provide Jira credentials plus either a single toolCall or a batch of toolCalls.

{
    "domain": "yourcompany",
    "email": "you@company.com",
    "apiToken": "your-jira-api-token",
    "toolCall": {
        "name": "jira_my_issues",
        "arguments": {}
    }
}

Run several tools in one pass with toolCalls (executed in order):

{
    "domain": "yourcompany",
    "email": "you@company.com",
    "apiToken": "your-jira-api-token",
    "toolCalls": [
        { "name": "jira_list_projects", "arguments": {} },
        { "name": "jira_search", "arguments": { "jql": "project = ENG AND status = 'In Progress'" } }
    ]
}

Input fields

Field	Required	Description
domain	For all live tools	Atlassian domain (e.g. yourcompany from yourcompany.atlassian.net)
email	For all live tools	Jira account email
apiToken	For all live tools	Jira API token (create at https://id.atlassian.com/manage-profile/security/api-tokens) — stored as a secret
toolCall	Optional	A single tool to run: { "name": ..., "arguments": {...} }. Prefilled with jira_info
toolCalls	Optional	An array of tool calls run in sequence; each item is { "name": ..., "arguments": {...} }

If no credentials are supplied, only jira_info returns data; every other tool returns an authentication error. If neither toolCall nor toolCalls is supplied, the run still succeeds and the dataset contains only the server-info and session-stats records.

Tools (14)

Tool	Description	Arguments (defaults)
jira_search	Search issues with JQL	jql (required), max_results (20), fields (default field set)
jira_get_issue	Get full issue details by key	key (required, e.g. PROJ-123)
jira_create_issue	Create a new issue	project, summary (required), issue_type (Task), description, priority, assignee_id, labels
jira_update_issue	Update issue fields	key (required), fields (object of fields to set)
jira_transition	Transition issue status by name	key, transition_name (required)
jira_add_comment	Add a plain-text comment	key, body (required)
jira_get_comments	Get comments on an issue	key (required), max_results (20)
jira_assign	Assign / unassign an issue	key (required), account_id (null = unassign)
jira_list_projects	List accessible projects	max_results (50)
jira_list_sprints	List sprints for a board	board_id (required), state (active)
jira_sprint_issues	Get issues in a sprint	sprint_id (required), max_results (50)
jira_my_issues	Get issues assigned to you	status (none), max_results (20)
jira_list_transitions	List available transitions for an issue	key (required)
jira_info	Server info, capabilities, and session stats	none

Output

Results are written to the Actor's default dataset, one record per event:

• mcp_server_info — server name/version, the tool catalog, and whether credentials were supplied
• tool_result — one per executed tool, with tool, arguments, and a result object (result.success plus the tool payload, or result.error on failure)
• session_stats — operation counters for the run (total_operations, reads, writes, errors)
• error — only if an unexpected run-level error occurred

The dataset "Overview" view surfaces type, tool, result.success, and result.key.

Authentication

Parameter	Description
domain	Your Atlassian domain (e.g. yourcompany from yourcompany.atlassian.net)
email	Your Jira account email
apiToken	Jira API token (create at https://id.atlassian.com/manage-profile/security/api-tokens)

Limitations

• Jira Cloud only. Uses https://<domain>.atlassian.net with REST API v3; Jira Server / Data Center is not supported.
• Credentials required. All tools except jira_info require domain, email, and apiToken. Only Atlassian Basic auth (email + API token) is supported — no OAuth. The account's own Jira permissions govern what each tool can do (e.g. creating, editing, transitioning, or commenting).
• Batch Actor, not a live MCP endpoint. Each run executes the provided toolCall/toolCalls and exits. The Actor does not run a persistent MCP transport (no stdio/SSE standby server), so AI assistants integrate by invoking the Actor per tool call rather than connecting to a streaming server.
• No pagination. Only the first page of results (up to max_results) is returned. Defaults: 20 for search/comments/my-issues, 50 for projects/sprint-issues. Large result sets are truncated.
• Plain text only. Descriptions and comments are sent and parsed as plain text via Atlassian Document Format. Rich formatting, @mentions, attachments, and embedded media are not supported.
• Sprint tools need Jira Software. jira_list_sprints and jira_sprint_issues require an Agile board; sprint/story-point fields depend on your Jira configuration (e.g. story points read from customfield_10016).
• No delete operations. The Actor can create and update issues but cannot delete them.
• Rate limits. Calls are subject to Atlassian's API rate limits; heavy batches via toolCalls may be throttled by Jira.
• Errors are reported, not fatal. Bad credentials, missing permissions, or a Jira/network outage are returned in the record's result.error field with success: false; the Actor run itself still completes.

Pricing

This Actor runs on the Apify platform and is billed according to the platform resources it consumes (compute units) for the duration of each run. See the Apify pricing page for current rates.