# How to connect Atlassian to Grok Bot

```json
{
  "title": "How to connect Atlassian to Grok Bot",
  "toolkit": "Atlassian MCP",
  "toolkit_slug": "atlassian_mcp",
  "framework": "Grok Bot",
  "framework_slug": "grok-bot",
  "url": "https://composio.dev/toolkits/atlassian_mcp/framework/grok-bot",
  "markdown_url": "https://composio.dev/toolkits/atlassian_mcp/framework/grok-bot.md",
  "updated_at": "2026-11-15T06:00:54.415Z"
}
```

## Introduction

Grok is xAI's assistant, and its bot builder lets you create a personal agent that holds context, runs skills, and connects to the apps you already use through plugins.
This guide explains how to connect Atlassian to a Grok bot using Composio Connect, which handles OAuth, token refresh, and reliability so you do not have to. You add Composio as a plugin, authenticate once, then ask your bot to work with Atlassian in plain language.
For exact tool slugs and parameter tables, see the [tool catalog in the Composio docs](https://docs.composio.dev/toolkits/atlassian_mcp).

## Also integrate Atlassian MCP with

- [ChatGPT](https://composio.dev/toolkits/atlassian_mcp/framework/chatgpt)
- [Claude Cowork](https://composio.dev/toolkits/atlassian_mcp/framework/claude-cowork)
- [Hermes](https://composio.dev/toolkits/atlassian_mcp/framework/hermes-agent)
- [Atomic Agent](https://composio.dev/toolkits/atlassian_mcp/framework/atomic-agent)

## TL;DR

Composio exposes 1,700+ apps to your Grok bot through a single connection, so you get:
- Connect to 1,700+ apps with on-demand tool loading, so you do not fill your context window with unnecessary tool definitions.
- Use programmatic tool calling through a remote workbench, letting the model write its own code to handle complex tool chaining. This reduces back-and-forth for frequent tool calls.
- Handle large tool responses outside the model context to keep conversations lean.

## Connect Atlassian MCP to Grok Bot

### Connecting Atlassian to Grok Bot
Once Grok Bot is set up, you connect Atlassian entirely from its interface, and the whole flow takes under a minute.
### Set up Grok Bot
Get access. Grok Bot is included with Cursor Ultra, Cursor Premium Teams, and SuperGrok Heavy. Get the desktop app for macOS or Windows from [x.ai/bot](https://x.ai/bot) and sign in. An iPhone companion app is also available, and your bots sync across devices.
Create a bot. Pick a suggested role or define your own, then give it a short name, one primary job, and a short description of how it should work. Focused bots work best, because a clear role gives the bot more useful context over time.
### Connect Atlassian
### 1. Add the Composio plugin
Click the button. The Composio listing on x.ai opens and launches Grok Bot for you. If the app does not open on its own, click Add to Grok Bot.
[Add to Grok Bot](https://x.ai/bot/plugin/32661537)
Grok Bot opens the Composio plugin, showing its connector and skills. Click Add.
### 2. Authenticate
Go to the Installed section and click the Composio entry. You'll see its connector and skills listed.
Click Authenticate. You'll be taken to the Composio consent screen. Review the access it requests and click Authorize. Composio handles the OAuth handshake for you, so you never touch API keys or tokens.
### 3. Connect your Atlassian account
Back in the chat, ask your bot to connect to Atlassian or give it any Atlassian task.
For example, ask it to:
- "Create a Jira issue for outage"
- "List open Jira issues assigned to me"
- "Create Bitbucket pull request from branch"
It will prompt you to authenticate and authorize access.
That's it. Composio's tools are now available in your Grok bot, and your Atlassian account is ready to use.

## What is Grok Bot?

Grok is xAI's assistant, available as a chat interface and a bot builder for creating personal agents that run skills and connect to external apps.
With Composio Connect, your Grok bot can securely access apps like Atlassian through MCP without you sharing account credentials directly with the agent.

## What is the Atlassian MCP server, and what's possible with it?

The Atlassian MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Atlassian account. It provides structured and secure access to Jira projects and issues, Confluence spaces and pages, Compass components, and Teamwork Graph context, so your agent can search work, create and update issues, publish Confluence pages, manage Compass components, and explore related context on your behalf.
- Jira issue management: Have your agent create and edit Jira issues, add comments and worklogs, link related issues, and move work through available statuses.
- Jira and Confluence search: Direct your agent to find issues, pages, and linked content with Rovo Search, or run targeted JQL and CQL searches when needed.
- Confluence content and discussions: Let the agent create Confluence pages, browse spaces and child pages, read page content, and add inline or footer comments.
- Compass component tracking: Instruct your agent to create and retrieve Compass components, define custom fields, and connect related components.
- Connected work context: Have your agent explore relationships around Atlassian work, retrieve connected objects, and add relationships between items in the Teamwork Graph.

## Supported Tools

| Tool slug | Name | Description |
|---|---|---|
| `ATLASSIAN_MCP_ADD_CONFLUENCE_CONTENT_PERMISSIONS` | Addconfluencecontentpermissions | Adds one or more content-level permission grants to a single Confluence content entity. Existing grants for other principals are preserved; to replace the full set, use replaceConfluenceContentPermissions instead. These are Specific access grants: operation `read` = Can view; `update` = Can edit and already includes view. A principal has exactly ONE level, so assigning `read` to a principal who currently has `update` is a downgrade to view-only. Never include the same principal under both operations. Adding a grant does NOT restrict others, make content private, or change General access; use setConfluenceContentRestrictionState for that baseline. Private creation already gives the creator `update`. Do not add `read` for that creator, because update already includes view. Existing-state and self-lockout rules are enforced by the Confluence monolith. Each grant pairs an operation with a list of principals. Pass the grants via the `permissions` input, an array of objects shaped exactly like: [ { "operation": "read", "principals": [ { "principalType": "user", "principalId": "" }, { "principalType": "group", "principalId": "" } ] } ] Enums: operation is one of read \| update; principalType is one of user \| group. For a user, principalId is the Atlassian account ID; for a group, principalId is the group ID. |
| `ATLASSIAN_MCP_ADD_CONFLUENCE_REACTION` | Addconfluencereaction | Add emoji reaction to Confluence content |
| `ATLASSIAN_MCP_ADD_CONNECTIONS_TO_GOALS` | Addconnectionstogoals | Add one type of connection to one or more Atlassian Goals. |
| `ATLASSIAN_MCP_ADD_CONNECTIONS_TO_PROJECTS` | Addconnectionstoprojects | Add one type of connection to one or more Atlassian Projects. |
| `ATLASSIAN_MCP_ADD_GRAPH_CONTEXT` | Addgraphcontext | Adds a relationship between two entities in the Teamwork Graph (e.g. linking two Jira work items, marking one as blocking another, attaching a remote link, or connecting a Jira work item to an Atlas project or goal). |
| `ATLASSIAN_MCP_ADD_LABELS_TO_CONFLUENCE_CONTENT` | Addlabelstoconfluencecontent | Add labels to any Confluence content (page, blogpost, attachment, custom content) |
| `ATLASSIAN_MCP_ADD_OR_EDIT_JIRA_ISSUE_COMMENT` | Addoreditjiraissuecomment | Add a comment to a Jira issue, or edit an existing one: omit commentId to add, or pass one from listJiraIssueComments to edit. Returns the commentId, needed to edit or delete it later. When no file was appended through inlineFileId, it also returns the echoed body and its format (appliedContentFormat). |
| `ATLASSIAN_MCP_ADD_OR_EDIT_JIRA_ISSUE_WORKLOG` | Addoreditjiraissueworklog | Log time spent on a Jira issue, or edit an existing worklog: omit worklogId to add, or pass one from listJiraIssueWorklogs to edit, which changes only the fields you pass. A new worklog is always logged as the calling user; an edit keeps its original author. Jira reduces the issue's remaining estimate by the time logged (by the difference on an edit). To change the estimate itself, use editJiraIssue. Returns the worklog's id, needed to edit it later, with the stored timeSpent and started. |
| `ATLASSIAN_MCP_APPROVE_JSM_REQUEST_RESOLVER_PLANS` | Approvejsmrequestresolverplans | Approve the current Request Resolver plans for explicitly selected JSM issue keys after the user has reviewed their plans with getJsmRequestResolverPlans and confirmed approval. Approval applies to each request's current plan at execution time, which may have changed since retrieval. The caller must be the request's human assignee or delegator. Returns one result per input in order; approved means approval was accepted, not that execution is complete. Output fields include requests[].issueKey, requests[].status, and approver.displayName; use these paths when selecting responseFields. Duplicate keys for the same request are submitted once per call. This is not idempotent across calls. For an unknown outcome, refresh the plan before any further approval; never automatically retry approval. Use getJsmRequestResolverPlans to check subsequent execution status. Does not decide customer-portal approvals. |
| `ATLASSIAN_MCP_ARCHIVE_CONFLUENCE_CONTENT` | Archiveconfluencecontent | Archive a single Confluence content entity |
| `ATLASSIAN_MCP_ATLASSIAN_USER_INFO` | Atlassianuserinfo | Returns the accountId of the current user |
| `ATLASSIAN_MCP_COMPLETE_CONFLUENCE_TASK` | Completeconfluencetask | Mark a Confluence inline task as complete |
| `ATLASSIAN_MCP_CONVERT_CONFLUENCE_CONTENT_MODE` | Convertconfluencecontentmode | Convert a Confluence page or live doc between classic page mode and live-doc mode |
| `ATLASSIAN_MCP_CONVERT_JIRA_ISSUE_HIERARCHY` | Convertjiraissuehierarchy | Convert an existing Jira issue across hierarchy levels: a standard issue (Task, Bug, Story) into a sub-task under a parent, or a sub-task back into a standard issue. Mirrors Jira's Move / Convert-to-sub-task wizard: preview destination compatibility (mode=preview_step1), preview field retain/remap/drop (mode=preview_step2), then execute (mode=confirm). Do not use this for same-level type edits such as Bug → Story — use editJiraIssue for those. Task → Sub-task: always ask the user for parentIssueKey before calling. Do not search for, infer, or pick a parent. If they did not name a parent, ask them instead of calling this tool. The parent must be a standard issue. If this result has CHILD_ISSUES_WILL_REPARENT, those children will get the same new parent — ask the user to accept that, then confirm with acknowledgeChildReparent=true. If this result has SOURCE_HAS_SUBTASKS, the convert is blocked; tell the user that. If this result has MOVE_PERMISSION_REQUIRED, the user lacks Jira Bulk change or Move permission — tell them to convert in the Jira UI. Do not retry. Do not mention another project style. If preview execution.method is bulk-move, or confirm returns MOVE_CONFIRMATION_REQUIRED, this is Jira's Move wizard: list the destination project, statusMapping, and fieldPlan.dropped to the user and ask them to accept that status change and that those fields will be dropped. Do not convert until they accept, then retry with mode=confirm and acknowledgeMove=true. If they decline, stop. Sub-task → Task: no parent needed; confirm clears the parent. After convert, reply with this issue's key, new type, and parent. If appliedChanges.dropped is present, also name those dropped fields. If appliedChanges.status is present, also name that status change. Do not mention bulk-move permissions or project-style notes unless this result has MOVE_PERMISSION_REQUIRED. |
| `ATLASSIAN_MCP_COPY_CONFLUENCE_CONTENT` | Copyconfluencecontent | Copy a Confluence page, whiteboard, database, folder, or embed to a new parent |
| `ATLASSIAN_MCP_CREATE_ARTIFACTS_UPLOAD` | Createartifactsupload | Step 1 of 2 for uploading a new artifact or a new version of an existing artifact. Creates an upload and returns a pre-signed URL and instructions for transferring the file's bytes directly. After the upload completes, call publishArtifact (step 2) with the returned completionPayload.contentId and artifact metadata to finalize and get the share URL. This does not publish the artifact by itself. Requires code execution and HTTP upload capability. |
| `ATLASSIAN_MCP_CREATE_ARTIFACT_THUMBNAIL_UPLOAD` | Createartifactthumbnailupload | Create a thumbnail for an artifact using an image generated locally or supplied by the user. Use only when the user explicitly requests a thumbnail. Requires artifact ownership and applies to the latest version. Creates a thumbnail only when none exists for that version. Overwriting, replacing, or regenerating an existing thumbnail is not supported. For HTML and PDF, use a static 672-by-400 PNG (1.68:1), capturing rendered HTML and a representative animation frame, or the first PDF page. Returns a pre-signed URL and instructions for uploading the image. No publishArtifact call is needed. |
| `ATLASSIAN_MCP_CREATE_CAPACITY_PLAN_NON_PROJECT_WORK` | Createcapacityplannonprojectwork | Create a non-project work item in Individual Capacity Planning. "Non-project work" is a bucket of time that is not a Jira issue or an Atlas project, such as Leave, Run the Business (RtB), on-call, or training. USE WHEN the user asks to create a new non-project work item, or a category of work that does not already exist as a Jira issue or Atlas project. Search first with listCapacityPlanNonProjectWork to avoid creating a duplicate - the match is on name and is case-insensitive. Non-project work is owned by Capacity Planning and exists nowhere else in Atlassian, so it can only be created through this tool. A contributor must be supplied: the new work item is created together with a work association for that person, so it is immediately allocatable to them. Allocate additional people or set the effort per period afterwards with updateCapacityPlanAllocations, using the ARI returned here as workId. On success, nonProjectWork contains the created item's name and ari - use that ari as workId for updateCapacityPlanAllocations rather than fabricating one. |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_ATTACHMENT` | Createconfluenceattachment | Prepare a Confluence attachment upload and return a local curl command |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_COMMENT` | Createconfluencecomment | Create or reply to a comment on a Confluence page or blog post |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_CONTENT` | Createconfluencecontent | Create a new Confluence page, blog post, live doc, whiteboard, database, embed, smart link, or folder. Before choosing body.format or authoring a body, call getContentFormatGuide through execute with name=getContentFormatGuide and inputs={toolName}. Use createConfluencePage for pages, blog posts, and live docs; createConfluenceWhiteboard for whiteboards; or createConfluenceDatabase for databases. Space instructions describe how to author content in a Confluence space. Apply relevant guidance to the content you create or update. Follow this decision rule: 1. Before authoring, call getConfluenceSpace for the target space unless getConfluenceContent returned metadata.hasSpaceInstructions=false for content in that space. If getConfluenceSpace already returned for the same cloudId and space during the current task, reuse its spaceInstructions and do not call it again. 2. When the value is true or unknown and instructions have not been loaded, you must call getConfluenceSpace once before authoring through execute with name=getConfluenceSpace, top-level cloudId, and inputs={spaceIdOrKey}. 3. Apply any returned spaceInstructions. If a successful response omits them, no instructions are configured; author with defaults and do not block. After a successful getConfluenceSpace call, do not make a separate getConfluenceSpaceInstructions call. |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_INFOGRAPHIC_FOR_PAGE` | Createconfluenceinfographicforpage | Create an infographic in a page workflow. Use only when a Confluence page creation or editing skill or guided workflow selects it; never as a standalone tool. Returns a media reference for embedding, which may sync later. Before prompting, reuse the shared Remix guide (headed `## Remix embeds`) if its text is in your current context. Only if missing, call getContentFormatGuide with toolName=createConfluenceInfographicForPage and follow its Remix guidance. Generation and billing finish asynchronously; typical cost is 15–30 Rovo credits. Download detailed billing data for final usage; match any returned correlationId and usageUniqueId. |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_MAUI_APP` | Createconfluencemauiapp | Create a Maui app in a page workflow. Use only when a Confluence page creation or editing skill or guided workflow selects it; never as a standalone tool. Returns a reserved app ID and embed URL for the owning page. Before prompting, reuse the shared Remix guide (headed `## Remix embeds`) if its text is in your current context. Only if missing, call getContentFormatGuide with toolName=createConfluenceMauiApp and follow its Remix guidance. Generation and billing finish asynchronously; typical cost is 15–30 Rovo credits. Download detailed billing data for final usage; match any returned correlationId and usageUniqueId. |
| `ATLASSIAN_MCP_CREATE_CONFLUENCE_SPACE` | Createconfluencespace | Create a Confluence space. Successful responses include id, key, and homepageId (the homepage / sidebar root content ID). |
| `ATLASSIAN_MCP_CREATE_FOCUS_AREA` | Createfocusarea | Create a new Focus area in a workspace. Call getFocusAreaTypes first and use an enabled type UUID for focusAreaTypeId — do not fabricate it. Optionally set parentFocusAreaIdentifier (UUID, ARI, or Focus page URL); if set, the parent's type must be a higher hierarchy level than the type being created. Optionally set startDate and targetDate as { date, confidence } where date is YYYY-MM-DDTHH:mm (e.g. 2026-08-10T00:00; bare dates are not accepted) and confidence is DAY, MONTH, or QUARTER. Omit owner to default to the creating user. Returns the created focus area (identifiers, type, owner, parent, dates, status, health, timestamps). Use getFocusArea for expansions such as aboutContent, statusUpdates, or subFocusAreas. When passing both cloudId and a parent ARI, they must refer to the same site. |
| `ATLASSIAN_MCP_CREATE_FOCUS_AREA_STATUS_UPDATE` | Createfocusareastatusupdate | Create a status update on a focus area, or save it as a draft. A status update is a change to one or more of status, health, summary, target date. Changing status needs an id discovered first with getFocusArea and include.statusTransitions; health does not. Passing only focusAreaIdentifier still posts a visible, empty update when publishing, so never call this tool speculatively. Field-level behavior for status, health, summary, target date, and draft constraints is documented on the inputs. |
| `ATLASSIAN_MCP_CREATE_GOAL` | Creategoal | Create an Atlassian Goal or Success Measure using an enabled Goal Type. |
| `ATLASSIAN_MCP_CREATE_GOAL_UPDATE` | Creategoalupdate | Create a progress update for an existing Atlassian Goal. |
| `ATLASSIAN_MCP_CREATE_JIRA_BOARD` | Createjiraboard | Create a new company-managed Jira Software board (scrum or kanban) from an existing filter. Requires a filter ID the user can view. When location is omitted, the board is filed in the single company-managed project from the filter's JQL. Team-managed (TMP) locations are always rejected: TMP projects are 1:1 project to board (one built-in board at project creation, no way to add another). User locations are also rejected. The response includes a url that opens the board. The filter decides which issues the board shows; location only decides where the board is filed in navigation. |
| `ATLASSIAN_MCP_CREATE_JIRA_ISSUE` | Createjiraissue | Create a new Jira issue. Custom fields go in additional_fields: keyed by field name, values are resolved automatically, so no field lookup is needed first (user fields take accountIds); keyed by customfield_* ID, pass the value in Jira's format. A rejected create returns a repairHint and names the field at fault where there is one, with its allowedValues when known, so retry from that. If sprint assignment fails, the response still has the new issue's key and sprintAssignment.assigned: false — do not create it again. |
| `ATLASSIAN_MCP_CREATE_JIRA_ISSUE_LINK` | Createjiraissuelink | Create a link between two existing issues — blocks, duplicates, clones, relates. Hierarchy is not a link: a parent or epic is the issue's parent field. To link an issue to anything outside Jira — a page, a document, any URL — use createJiraIssueRemoteIssueLink. |
| `ATLASSIAN_MCP_CREATE_JIRA_ISSUE_REMOTE_ISSUE_LINK` | Createjiraissueremoteissuelink | Create a native web link in a Jira issue's Web links section for a Confluence page or external HTTP(S) URL. Not a URL custom field; for an issue-to-issue dependency use createJiraIssueLink. |
| `ATLASSIAN_MCP_CREATE_JSM_REQUEST` | Createjsmrequest | Create a Jira Service Management customer-request work item using an explicit JSM service desk ID, JSM request type ID, and request field values. Do not use Jira issue type IDs or generic Jira create-issue workflows. First use getJsmRequestCreationOptions to select a JSM request type, then getJsmRequestTypeFields to build requestFieldValues. |
| `ATLASSIAN_MCP_CREATE_LOOM_FOLDER` | Createloomfolder | Create a new folder in the current Loom workspace, optionally nested under a parent folder or placed in a space. Returns the new folder's id, which you can pass to moveLoomVideoToFolder. To list existing folders instead of creating one, use listLoomFolders. |
| `ATLASSIAN_MCP_CREATE_LOOM_VIDEO_COMMENT` | Createloomvideocomment | Post a comment on a Loom video, anchored to a specific point in the video. The timestamp is in SECONDS (for example 90 for the 1:30 mark). Use this to leave feedback or a note at a moment in the recording. To read existing comments use getLoomVideo with expand: ["comments"]; to add an emoji reaction instead of a text comment use createLoomVideoReaction. Identify the video by ARI or share URL. NOTE: the returned timestamp is remapped onto the video's current (post-trim) timeline, so for a video with trimmed/edited ranges the echoed value can come back up to 1 second lower than the value sent (never higher); for videos with no trims it always matches exactly. |
| `ATLASSIAN_MCP_CREATE_LOOM_VIDEO_REACTION` | Createloomvideoreaction | Add an emoji reaction to a Loom video, anchored to a specific point in the video. The timestamp is in SECONDS (for example 90 for the 1:30 mark), the same unit as createLoomVideoComment. Use a Loom-native reaction name (joy, love, wow, nervous, yay, up, down) or an extended emoji slug. To leave a text comment instead, use createLoomVideoComment. Identify the video by ARI or share URL. |
| `ATLASSIAN_MCP_CREATE_LOOM_VIDEO_UPLOAD` | Createloomvideoupload | Step 1 of 2 for uploading a new video to Loom. Creates an upload and returns a ready-to-run local script (plus raw upload endpoints) that transfers the file's bytes. After the script reports the upload is complete, call publishLoomVideo (step 2) with the returned completionPayload to finalize and get the share URL. This does not publish the video by itself. |
| `ATLASSIAN_MCP_CREATE_PROJECT` | Createproject | Create an Atlassian Project. |
| `ATLASSIAN_MCP_CREATE_PROJECT_UPDATE` | Createprojectupdate | Post a progress update on an existing Atlassian Project, or save it as an unposted draft with isDraft. Posting is immediately visible on the Project and notifies its followers, so only post when the user asked to; use isDraft to leave it for them to review and post themselves. The response carries isDraft so you can tell which happened, and a draft returns fewer fields than a posted update. |
| `ATLASSIAN_MCP_CREATE_TEAM` | Createteam | Create an Atlassian team with a bounded name, optional description, MEMBER_INVITE or OPEN membership settings, and up to 50 initial account IDs. Tenant and organization are derived by the server. |
| `ATLASSIAN_MCP_DECIDE_JSM_REQUEST_APPROVAL` | Decidejsmrequestapproval | Approve or decline a pending approval on an existing Jira Service Management request. Provide the request issue ID/key and an explicit decision of approve or decline. approvalId is optional; when omitted, the tool discovers pending request approvals and proceeds only when exactly one answerable approval is unambiguous. Use this for approval decisions, not workflow transitions, field edits, or comments. |
| `ATLASSIAN_MCP_DIFF_CONFLUENCE_CONTENT_VERSIONS` | Diffconfluencecontentversions | Compute a unified diff between two versions of a Confluence page, live doc, or blog post |
| `ATLASSIAN_MCP_DISABLE_CONFLUENCE_PUBLIC_LINK` | Disableconfluencepubliclink | Disable the external public link (anonymous share) for a Confluence page |
| `ATLASSIAN_MCP_DOWNLOAD_CONFLUENCE_ATTACHMENT` | Downloadconfluenceattachment | Get a Confluence attachment download URL and return a local curl command |
| `ATLASSIAN_MCP_DOWNLOAD_JIRA_ISSUE_ATTACHMENT` | Downloadjiraissueattachment | Get a short-lived download URL for a Jira attachment and return a local downloadCommand. Run it from your shell to save the file. |
| `ATLASSIAN_MCP_EDIT_CONFLUENCE_INFOGRAPHIC_FOR_PAGE` | Editconfluenceinfographicforpage | Edit an infographic in a page workflow. Use only when a Confluence page editing skill or guided workflow selects it; never as a standalone tool. Regenerates the existing infographic and returns a new placeholder media reference for embedding. Before prompting, reuse the shared Remix guide (headed `## Remix embeds`) if its text is in your current context. Only if missing, call getContentFormatGuide with toolName=editConfluenceInfographicForPage and follow its Remix guidance. Generation and billing finish asynchronously; typical cost is 15–30 Rovo credits. Download detailed billing data for final usage; match any returned correlationId and usageUniqueId. |
| `ATLASSIAN_MCP_EDIT_CONFLUENCE_MAUI_APP` | Editconfluencemauiapp | Edit a Maui app in a page workflow. Use only when a Confluence page editing skill or guided workflow selects it; never as a standalone tool. Regenerates the existing app and returns a new placeholder Maui ID and embed URL. Before prompting, reuse the shared Remix guide (headed `## Remix embeds`) if its text is in your current context. Only if missing, call getContentFormatGuide with toolName=editConfluenceMauiApp and follow its Remix guidance. Generation and billing finish asynchronously; typical cost is 15–30 Rovo credits. Download detailed billing data for final usage; match any returned correlationId and usageUniqueId. |
| `ATLASSIAN_MCP_EDIT_JIRA_ENTITY_PROPERTY` | Editjiraentityproperty | Sets a single property (technical key/value store) on a specific Jira entity (issue, issueComment, issueWorklog, dashboardItem, user), creating or overwriting it. Returns {key, entityType, status}, status being "created" or "updated"; the value you wrote is not returned — read it back with getJiraEntityProperty. |
| `ATLASSIAN_MCP_EDIT_JIRA_ISSUE` | Editjiraissue | Edit an existing Jira issue. Custom fields go in additional_fields: keyed by field name, values are resolved automatically (user fields take accountIds); keyed by customfield_* ID, pass the value in Jira's format. Every field is set, not appended: send the full list for labels, components and other multi-value fields. Change status or move to a sprint with transitionJiraIssue, and convert Task↔Sub-task with convertJiraIssueHierarchy. Empty description or environment with "". When editing a body that holds inline media, use contentFormat: html, or the media is dropped. Reorder on a board with rankBeforeIssue or rankAfterIssue; fields are saved first, so a failed reorder leaves the edit applied and returns ranking.ranked: false. The response shows only getJiraIssue's default fields, so other fields you set are not echoed back. |
| `ATLASSIAN_MCP_ENABLE_CONFLUENCE_PUBLIC_LINK` | Enableconfluencepubliclink | Enable the external public link (anonymous share) for a Confluence page |
| `ATLASSIAN_MCP_EXPORT_CONFLUENCE_CONTENT` | Exportconfluencecontent | Export Confluence content as PDF or Word and return a download link |
| `ATLASSIAN_MCP_GET_ACCESSIBLE_ATLASSIAN_RESOURCES` | Getaccessibleatlassianresources | Returns the cloudId(s) available to the current user |
| `ATLASSIAN_MCP_GET_ARTIFACT` | Getartifact | Get metadata for an artifact by ID. |
| `ATLASSIAN_MCP_GET_ASSETS_OBJECT` | Getassetsobject | Get an Assets object by ID from the selected site. |
| `ATLASSIAN_MCP_GET_ASSETS_OBJECTS_BY_ID` | Getassetsobjectsbyid | Get one or more Assets objects by ID from the selected site. |
| `ATLASSIAN_MCP_GET_ASSETS_OBJECT_SCHEMA` | Getassetsobjectschema | Get one or more Assets object schemas by ID from the selected site. |
| `ATLASSIAN_MCP_GET_ASSETS_OBJECT_TYPE` | Getassetsobjecttype | Get one or more Assets object types by ID from the selected site. |
| `ATLASSIAN_MCP_GET_ASSETS_SERVICE_OBJECTS_BY_ID` | Getassetsserviceobjectsbyid | Get one or more Assets service objects by ID from the selected site. |
| `ATLASSIAN_MCP_GET_CAPACITY_PLAN_ALLOCATIONS` | Getcapacityplanallocations | Find who is allocated to what work in Individual Capacity Planning, across ALL capacity plans - not scoped to a single plan. USE WHEN the user asks what someone is working on, who is working on an issue or project, whether a contributor is on a given work item, how loaded or available someone is, or how staffed a work item is. People are contributors, not assignees: allocation is independent of the Jira assignee field - use the Jira tools for assignees. Pass contributorDataIds for the work those people are allocated to, workDataIds for who is allocated to those items, or both to check one contributor against one item. At least one is required; resolve non-project work ARIs with listCapacityPlanNonProjectWork rather than fabricating them. Returns contributor-work associations with contributor and work names/keys. expand adds 'contributions' (allocated value per period), 'contributorAggregations' (a contributor's total contribution AND availability across all plans - "how loaded is X") and 'workAggregations' (total effort across all contributors on an item - "how staffed is PROJ-42"); both aggregations come back as separate top-level objects, and all three can be combined. IMPORTANT: contributionStartDate and contributionEndDate do NOT filter which associations are returned - they only set the window for the contribution and aggregation figures. To check whether a contributor is allocated during a period, request contributions for that window and see whether any come back. A baseline 40-hour working week means 5 days, 40 hours and 100 percent are the same amount - say so when reporting days or percentages. PAGINATION - YOU MUST CHECK THIS: results come back one page at a time. When the response has pageInfo.hasNextPage = true the allocations list is INCOMPLETE - more contributor-work associations exist that are not in this response. Do NOT draw conclusions (totals, counts, "that's the whole team", "nobody else is allocated") from a single page when hasNextPage is true. To get the rest, call again with after set to pageInfo.endCursor and repeat until hasNextPage = false, then combine every page before answering. For a broad "everyone's allocations for this range" request, keep paging until hasNextPage = false. |
| `ATLASSIAN_MCP_GET_CODE_FILE` | Getcodefile | Read the full contents of a single file in a connected source-control repository at a specific commit. Use this to inspect a file surfaced by searchCode: pass matchedFiles[].workspace, repo, path, and version (as commitHash) from that result. Returns the file's text and a link to it. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_APPROVAL` | Getconfluenceapproval | Get Confluence page approval availability, approval status, reviewers, and optional progress or history. Omit contentId to list the pages awaiting the current user's approval decision. For a general page status request, also call getConfluenceContentStatus; for approval status, use this tool only. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_ATTACHMENT` | Getconfluenceattachment | Get Confluence attachment metadata by attachment ID |
| `ATLASSIAN_MCP_GET_CONFLUENCE_COMMENT` | Getconfluencecomment | Get a single Confluence comment by ID |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT` | Getconfluencecontent | REQUIRED: Supply at least one content locator: content_id or content_url. Read any Confluence content (page, blog, live doc, comment, whiteboard, embed, database, folder). |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_ANALYTICS` | Getconfluencecontentanalytics | Return per-user Confluence viewer rows for one supported content item (page, live doc, or blog post), including each viewer's engagement category, the latest version viewed, and last viewed time. Always present results as a horizontal table, even for one viewer, with exactly these columns in this order: People and agents, Engagement, Latest version viewed, Last viewed, Views. When dates are omitted, Confluence uses the same default range as the Viewers table. Use the relative last-viewed value, such as "7 minutes ago", instead of the exact timestamp. For an AGENT row with both names, present People and agents as " with ". Keep the Engagement column; if engagement is missing, put "-" rather than inventing another category. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_ANCESTORS` | Getconfluencecontentancestors | List the ancestor chain (parent, grandparent, …) of a Confluence content entity |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_DESCENDANTS` | Getconfluencecontentdescendants | List descendants of a Confluence content entity (page, folder, whiteboard, database, embed, or slide) |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_PERMISSIONS` | Getconfluencecontentpermissions | Lists the explicit content-level permission grants on a single Confluence content entity. Each grant pairs an operation (read or update) with a principal (a user or a group). Optionally filter by operation, principal-type, or principal-id, and paginate with cursor and limit. Use this to audit who has been granted read or update access before adding, replacing, or removing grants. To read the higher-level restriction state (OPEN / EDIT_RESTRICTED / VIEW_RESTRICTED), use getConfluenceContentRestrictionState instead. Specific access rows: `read` = Can view and `update` = Can edit (which already includes view). Each principal has one level: read, update, or none. These grants add principal access; they do not restrict other people or change General access. General access is controlled by the restriction state, not direct grants. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_RESTRICTION_STATE` | Getconfluencecontentrestrictionstate | Reads General access for a single Confluence content entity. The restriction states are: OPEN — "Open, Anyone in this space can edit". EDIT_RESTRICTED — "Open, Anyone in this space can view"; only Specific access update principals can edit. VIEW_RESTRICTED — "Restricted, Only specific people can view or edit"; Specific access read/update grants determine access. Use this for a quick read of the overall restriction posture; use getConfluenceContentPermissions to enumerate the individual user/group grants. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_STATUS` | Getconfluencecontentstatus | Read the content state (status) of a Confluence page or blog post |
| `ATLASSIAN_MCP_GET_CONFLUENCE_CONTENT_VERSION` | Getconfluencecontentversion | Get one historical version of a Confluence page, live doc, or blog post |
| `ATLASSIAN_MCP_GET_CONFLUENCE_MAUI_APP` | Getconfluencemauiapp | Retrieve the source and metadata for a Maui app owned by a Confluence page. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_PERSONAL_SPACE` | Getconfluencepersonalspace | Get the current user's Confluence personal space |
| `ATLASSIAN_MCP_GET_CONFLUENCE_PUBLIC_LINK_STATUS` | Getconfluencepubliclinkstatus | Get the external public-link (anonymous share) status of a Confluence page |
| `ATLASSIAN_MCP_GET_CONFLUENCE_REACTIONS` | Getconfluencereactions | Get emoji reactions on Confluence content |
| `ATLASSIAN_MCP_GET_CONFLUENCE_SPACE` | Getconfluencespace | Get one Confluence space by id or key, including enabled content types, optional labels, and instructions when configured. Reuse the result for the same cloudId and space during the current task instead of reading the space before every write. If spaceInstructions is omitted, no instructions are configured; normal authoring does not require a separate getConfluenceSpaceInstructions call. |
| `ATLASSIAN_MCP_GET_CONFLUENCE_SPACE_INSTRUCTIONS` | Getconfluencespaceinstructions | Get the space instructions for a Confluence space |
| `ATLASSIAN_MCP_GET_CONFLUENCE_TASK` | Getconfluencetask | Get a single Confluence inline task by its ID |
| `ATLASSIAN_MCP_GET_CONFLUENCE_TEMPLATE` | Getconfluencetemplate | Get a Confluence template body |
| `ATLASSIAN_MCP_GET_CONTENT_FORMAT_GUIDE` | Getcontentformatguide | Get the content-format guide for authoring a rich content body or constructing a Confluence Remix prompt. Pass `toolName` to receive the canonical guidance for that operation. For generic Confluence content writes, pass the content-specific key (for example `createConfluencePage`, not `createConfluenceContent`). For Remix, pass the exact create/edit operation you are about to call. Use this before authoring or editing rich Confluence content, constructing a Remix generation/edit prompt, or writing a Jira issue/comment body in HTML. First call with only `toolName`; if the returned guidance references follow-up documents, call this operation again with the same `toolName` and `reference` set to a single referenced path to load that additional guidance. Fetch one document per call. |
| `ATLASSIAN_MCP_GET_FOCUS_AREA` | Getfocusarea | Get a focus area by identifier, with optional expansions. Use when the exact focus area UUID, ARI, or page URL is known. For text search or listing, use searchFocusAreas instead. Optional include expansions cover about content, the current user's latest draft status update, status updates, sub-focus areas, linked goals, linked work, and custom field values. |
| `ATLASSIAN_MCP_GET_FOCUS_AREA_TYPES` | Getfocusareatypes | Get the configured Focus area type hierarchy for a workspace. USE WHEN the user needs the valid Focus area types for a site — for example before creating a Focus area, or to understand hierarchy levels and parent type relationships. Returns each type with identifiers, hierarchy level, parent hierarchy references, and associated Focus area counts. |
| `ATLASSIAN_MCP_GET_GOAL` | Getgoal | Retrieve one Atlassian Goal by its exact Goal ARI or key, with optional related details. |
| `ATLASSIAN_MCP_GET_GOAL_TYPES` | Getgoaltypes | Get the Goal Types available in an Atlassian site for creating Goals and Success Measures. |
| `ATLASSIAN_MCP_GET_GOAL_UPDATE` | Getgoalupdate | Retrieve one Atlassian Goal update by its exact Goal Update ARI, with optional related details. |
| `ATLASSIAN_MCP_GET_GRAPH_CONTEXT` | Getgraphcontext | Retrieves connected context from Teamwork Graph for any Atlassian entity. Returns all relationships and linked objects within one traversal – including cross-product and third-party connections. Use when the answer requires connections between entities, not just a single entity's fields. Supported entry points: Jira: issues, projects, sprints, versions, comments Confluence: pages, blogposts, whiteboards, databases, spaces Goals, Projects, Focus Areas, Tags and updates People: users, teams, organisations DevOps: PRs, repos, deployments, services, builds, designs Loom: videos, meetings Compass: components Assets: objects Incidents, conversations, calendar events, external documents After calling this tool, call getGraphObject on returned ARIs to get full content. The utilisation of this tool may consume up to 10 Rovo Credits. If the user or organization does not have any Rovo Credits available, the tool call will fail. |
| `ATLASSIAN_MCP_GET_GRAPH_OBJECT` | Getgraphobject | Fetches the entire available data for one or more objects (Atlassian or third-party) using their ARIs or URLs. Use for the objects gathered from getGraphContext tool. |
| `ATLASSIAN_MCP_GET_JIRA_BOARD_CONFIG` | Getjiraboardconfig | Returns the configuration for a Jira board; filter, column layout, and each column's mapped statuses with their id, name and category. For the issues on the board, use getJiraBoardIssueData. |
| `ATLASSIAN_MCP_GET_JIRA_BOARD_ISSUE_DATA` | Getjiraboardissuedata | Get the issues on a Jira board, its backlog, or all of them, with optional count and JQL filtering. For an issue search that is not scoped to a board, use searchJiraIssuesUsingJql. For one sprint's issues with progress metrics, use getJiraBoardSprintData — on a Scrum board dataset "board" spans current and future sprints, not just the active one. |
| `ATLASSIAN_MCP_GET_JIRA_BOARD_SPRINT_DATA` | Getjiraboardsprintdata | Resolve a Jira board plus its sprint data in one call. Supports board lookup by name, sprint lookup by name, and defaults to the active sprint when sprintId and sprintName are omitted. Returns the board, the resolved sprint, all sprint metadata on that board, and — unless includeIssues is false — the sprint's issues with computed focus and progress slices such as assigned-to-me and at-risk items. If the requested sprint name or ID is not found, the tool returns the resolved board plus the available sprint names to help the agent repair the request. To only list a board's sprints or filter them by state, use listJiraBoardSprints. For board or backlog issues that are not scoped to a single sprint, use getJiraBoardIssueData. |
| `ATLASSIAN_MCP_GET_JIRA_CURRENT_USER` | Getjiracurrentuser | Get the calling user's own Jira profile. Use it to turn "me" or "my" into an accountId, and to read the user's time zone before resolving relative dates. For a JQL search, skip this tool and write assignee = currentUser() in the query instead. For anyone else use lookupJiraAccountId or getJiraUser. |
| `ATLASSIAN_MCP_GET_JIRA_ENTITY_PROPERTY` | Getjiraentityproperty | Returns a single property (technical key/value store) from a specific Jira entity (issue, issueComment, issueWorklog, dashboardItem, user), or lists all its property keys when propertyKey is omitted. With propertyKey it returns {key, value} with the stored JSON, null-valued fields generally dropped except assignee; without one, {keys: [...]} — key names only, no values. |
| `ATLASSIAN_MCP_GET_JIRA_ISSUE` | Getjiraissue | Get a Jira or JSM issue by key or ID. Use for one known issue; use searchJiraIssuesUsingJql to find or list many issues. Reports how many comments an issue has, not the comments themselves — for those use listJiraIssueComments. View "evidence" also returns the issue's links to other issues. |
| `ATLASSIAN_MCP_GET_JIRA_ISSUE_TYPE_META_WITH_FIELDS` | Getjiraissuetypemetawithfields | Returns the schema and field metadata for a specific issue type in a project, useful for creating issues with correct fields. Get issueTypeId from listJiraProjectIssueTypesMetadata first. Each entry in fields gives the fieldId, its name, whether it is required, schema.type — with schema.items naming the element type on arrays — and allowedValues on fields whose value must come from a predefined list. Only required fields are returned unless you pass requiredFieldsOnly: false. |
| `ATLASSIAN_MCP_GET_JIRA_PROJECT_VERSION_RELATED_WORK` | Getjiraprojectversionrelatedwork | List the related work on a Jira project version (release): the designs, test plans, runbooks and change records teams attach to it. Use an item's relatedWorkId to edit it with manageJiraProjectVersionRelatedWork; Jira's own release-note item has no relatedWorkId and cannot be edited. |
| `ATLASSIAN_MCP_GET_JIRA_PROJECT_VERSIONS` | Getjiraprojectversions | List a Jira project's releases/versions (pass projectIdOrKey), or fetch a single version (pass versionId) — provide exactly one. Results come back in a paginated envelope (values[]); in single-version mode the list-only inputs (startAt, maxResults, query, status, orderBy) are ignored. Optionally set include to add the "driver" and/or "approvers" to every version returned, but leave it unset unless the user explicitly asks for them. To create, update, or release a version, use manageJiraProjectVersion instead. |
| `ATLASSIAN_MCP_GET_JIRA_USER` | Getjirauser | Get a Jira user's profile by account ID, including time zone, locale and active status, plus groups and application roles when asked for via expand. If you do not have the accountId, use lookupJiraAccountId or listJiraIssueAssignableUsers first. For the calling user's own profile use getJiraCurrentUser instead. |
| `ATLASSIAN_MCP_GET_JSM_AGENT_AVAILABILITY` | Getjsmagentavailability | Read Jira Service Management Workforce Management availability for agents in a service project. Workforce Management uses agent availability alongside capacity, skills, and shift coverage when routing work. Returns project-level availability counts by default. Set includeSelf to read the authenticated agent's project-scoped status, or set includeAgents to true to list individual agents, statuses, teams, or paginated results. This is read-only; it does not assign work or change availability. Output fields include summary for project counts, agents[] for individual agents, and myAvailability for includeSelf; use those names when selecting responseFields. |
| `ATLASSIAN_MCP_GET_JSM_OPS_ALERTS` | Getjsmopsalerts | Get one Jira Service Management Operations alert by ID or alias, or list and search alerts with optional query, time window, and pagination. If multiple selectors are supplied, id takes precedence, then alias, otherwise list/search inputs are used. |
| `ATLASSIAN_MCP_GET_JSM_OPS_SCHEDULE_INFO` | Getjsmopsscheduleinfo | Get Jira Service Management Operations schedule information: list schedules, get a specific schedule by ID, or retrieve current or next on-call responders for a schedule. Use action="schedule" to list schedules or get schedule details. Use action="oncall" with scheduleId to get on-call responders. |
| `ATLASSIAN_MCP_GET_JSM_OPS_TEAM_INFO` | Getjsmopsteaminfo | Get Jira Service Management Operations team information: list all teams or get escalation policies for a specific team. Use action="teams" to retrieve a list of all teams based on the user's role. Use action="escalations" to list escalation policies for a team and obtain escalationId values used by updateJsmOpsAlert. |
| `ATLASSIAN_MCP_GET_JSM_REQUEST` | Getjsmrequest | Get details and the current customer-facing state of an existing Jira Service Management (JSM) customer request by issue key or ID. Returns requestType (request type), serviceDesk (service desk), reporter (requester), requestParticipants (participants), requestFields (request fields), status, agentUrl (request link), and nextActions. Use these exact output names when selecting responseFields. Supports summary and detail views. For editable fields on the underlying Jira issue, call getJiraIssue with includeEditableFields=true. Use this read-only tool to understand a request; use request action-metadata tools for transitions, editable fields, and valid options. Approvals are returned only when explicitly requested. |
| `ATLASSIAN_MCP_GET_JSM_REQUEST_ACTION_METADATA` | Getjsmrequestactionmetadata | Discover customer-visible request-portal actions for an existing Jira Service Management request before a follow-up write. Returns compact customer transitions by default and can explicitly include request type fields metadata or approvals. Uses JSM customer transitions, not Jira workflow transitions, and never mutates the request. The output field is customerTransitions (an array of id, name, and description); select customerTransitions, not transitions.*, when using responseFields. |
| `ATLASSIAN_MCP_GET_JSM_REQUEST_CREATION_OPTIONS` | Getjsmrequestcreationoptions | Prepare creation of a Jira Service Management customer-request work item for a project. Use this first to return its service desk and available request types. Select a request type, then use getJsmRequestTypeFields before createJsmRequest. Requires projectKey, the key of an accessible JSM service desk project (for example ITSM), not a numeric project ID. JSM request types are not Jira issue types. |
| `ATLASSIAN_MCP_GET_JSM_REQUEST_RESOLVER_PLANS` | Getjsmrequestresolverplans | Retrieve a concise snapshot of Request Resolver plans for selected Jira Service Management requests by issue key or ID. Returns one result per distinct input in first-selection order, containing loading state, plan status, readable steps, runbook links or an error. Ready plans remain available when other requests are loading or fail. Excludes internal IDs, sources and reasoning. isLoading can mean generating or executing; a missing plan alone does not imply loading. Requires Jira read access to resolve ticket and project IDs. Does not approve or execute plans. |
| `ATLASSIAN_MCP_GET_JSM_REQUEST_TYPE_FIELDS` | Getjsmrequesttypefields | Get field IDs, required flags, and allowed values for selected Jira Service Management work-item request type. Use getJsmRequestCreationOptions first, then build createJsmRequest.requestFieldValues from these fields. |
| `ATLASSIAN_MCP_GET_LOOM_VIDEO` | Getloomvideo | Reads a Loom video's core metadata and optionally its transcript, comments, AI agent briefs, AI-extracted meeting action items, or attached links in one call via the expand input. USE WHEN the user asks about a specific Loom video — its title, description, duration, owner, view/comment/reaction counts, what was said in it (pass expand: ["transcript"]), what feedback or comments people left on it (pass expand: ["comments"]), what the AI summarised as key points (pass expand: ["agentBriefs"]), or what follow-up tasks or action items the AI extracted from a meeting recording (pass expand: ["meetingActionItems"]), or which links are attached to it (pass expand: ["links"]). DO NOT USE to get the downloadable MP4 file — use getLoomVideoDownloadUrl for that. To find a video's ARI first, use listLoomVideos or listLoomVideosSharedWithMe. |
| `ATLASSIAN_MCP_GET_LOOM_VIDEO_DOWNLOAD_URL` | Getloomvideodownloadurl | Get a time-limited signed URL to download the actual MP4 file of a Loom video. This is the tool to use when the user wants the video file itself (to download, re-upload, or process), not its transcript or metadata (use getLoomVideo for those). MP4 generation is asynchronous: the response includes a status and, when not yet ready, a retryAfterSeconds hint — poll this tool until status is READY, then use the returned url before expiresAt. Identify the video by ARI or share URL. |
| `ATLASSIAN_MCP_GET_MY_WORK` | Getmywork | Read the signed-in user's recent Jira, Confluence, Goals, Projects and Loom activity on one site. Returns things worked on or viewed, newest activity first, with titles, URLs and activity timestamps. Uses Activity Platform with no enriched graph-context or Cypher calls. Assigned work is not part of this feed. The lookback is configurable and defaults to 90 days. |
| `ATLASSIAN_MCP_GET_PROJECT` | Getproject | Retrieve one Atlassian Project by its exact Project ARI or key, with optional related details. |
| `ATLASSIAN_MCP_GET_PROJECT_UPDATE` | Getprojectupdate | Retrieve one Atlassian Project update by its exact Project Update ARI, with optional related details. |
| `ATLASSIAN_MCP_GET_TALENT_FIELDS` | Gettalentfields | Discover the workspace's queryable Talent fields, or resolve the valid filter values for a specific field. Call BEFORE authoring RQL: with no fieldId it lists the available field IDs (including custom fields), grouped by what they describe (position, worker, team, skill, focus area, focus area type, proposal, proposed movement and view fields), excluding hidden fields; with a fieldId it returns that field's valid values, paginated. Field IDs and values are workspace-specific and must not be guessed. |
| `ATLASSIAN_MCP_GET_TALENT_GROUP_METRICS` | Gettalentgroupmetrics | Aggregate Talent positions into COUNTS grouped by one to four position field IDs — headcount by country, by status, by level, by job family, by manager, or any nesting of those. Returns only the grouped counts and their field values, never the positions themselves: when the underlying list is wanted use searchTalentPositions, and when the breakdown is by reporting line or focus area use getTalentPositionsByEntity, which groups by entity rather than by field. Grouping field IDs are workspace-specific, must be groupable, and must not be guessed — call getTalentFields to resolve them. The common built-ins are positionCountry, positionStatus, positionLevel, positionRole, positionJobFamily, positionJobTitle, positionEmploymentType, positionManager and team. |
| `ATLASSIAN_MCP_GET_TALENT_POSITION` | Gettalentposition | Read a single Talent position by its ARI, returning its selected position field values and, optionally, field values for the worker filling it. Use this only when you already have the position ARI: take it from a searchTalentPositions result's id, or from a getTalentPositionsByEntity result's entityId — NOT that tool's id, which is a connection node identifier and is not accepted here. To FIND a position rather than read a known one, use searchTalentPositions. To identify the worker filling the position, request positionWorker in positionFieldIdIsIn — that yields the worker's id, name and accountId. Do not rely on workerFieldIdIsIn for more than that: the backing action accepts it but does not currently return the requested worker field values, so attributes like skills or email will not come back here. Field IDs are workspace-specific and must not be guessed — call getTalentFields to resolve them. Common position built-ins are positionJobTitle, positionCountry, positionLevel, positionStatus, positionManager and positionWorker. |
| `ATLASSIAN_MCP_GET_TALENT_POSITIONS_BY_ENTITY` | Gettalentpositionsbyentity | List the Talent positions attached to an entity — the direct reports of a manager position (entity=position with that manager's ARI as fieldValue), or the positions rolled up to a focus area (entity=focusArea). This is the tool for walking the org chart and for breaking staffing down across focus areas or reporting lines, because it is the only one that can scope to those entities; it returns the underlying positions, paginated forward with after + endCursor. The two entities are scoped by DIFFERENT inputs: entity=position scopes with fieldValue, entity=focusArea scopes with rql. fieldValue is ignored when entity=focusArea, so a focus area passed there returns every row unfiltered — use rql "focusArea = ''" for that one area, or rql "focusAreaParent = ''" for its direct children. Use getTalentGroupMetrics INSTEAD when you want counts grouped by a position FIELD and not the positions themselves, and searchTalentPositions INSTEAD for a flat attribute filter that is not entity-scoped. Each returned position carries both an entityId and an id: entityId is the position identifier to pass to getTalentPosition, while id identifies the row within this connection and is NOT accepted as a position ARI. Field IDs and RQL values are workspace-specific and must not be guessed — resolve them with getTalentFields; focusArea and focusAreaParent are standard field IDs and are safe to use exactly as written. Useful built-ins are positionWorker, team, positionJobTitle, positionRole, positionStatus, aiMonthlyUsageCostSelfTotal and aiMonthlyUsageTokenSelfTotal. |
| `ATLASSIAN_MCP_GET_TALENT_USER` | Gettalentuser | Resolve the calling user's own place in the Talent workspace: their position entity ID, their manager's position entity ID, and their role. Call this FIRST to anchor any self-referential request ("my team", "my reports", "who do I report to") before using the other Talent tools. Note this returns entity IDs, not position ARIs — to get a position ARI for getTalentPosition or getTalentPositionsByEntity, look the position up with searchTalentPositions. |
| `ATLASSIAN_MCP_GET_TEAM` | Getteam | Get core metadata for one Atlassian team. Optionally expand any unique subset of members and hierarchy; each expansion supports its own cursor. |
| `ATLASSIAN_MCP_LINK_TO_FOCUS_AREA` | Linktofocusarea | Link one of either goals, work items, or sub-focus areas to a focus area. Add exactly one connection type per call. Prefer identifiers from a prior lookup, or call a lookup to get them; do not fabricate ARIs. Useful lookup helpers: getGoal for goal ARIs, getJiraIssue for Jira work ARIs, and getProject for Atlassian project ARIs. In this tool, WORK_ITEM means one of the supported work ARI families: Jira/JPD issues or Atlassian projects. focusAreaIdentifier accepts a focus area UUID, ARI, or page URL. Linking GOAL or WORK_ITEM requires focusAreaIdentifier to be a focus area ARI; if only a UUID or URL is known, call getFocusArea and pass the returned ari. connection must specify exactly one type: GOAL with goalAris, WORK_ITEM with workAris, or FOCUS_AREA with focusAreaIdentifiers. |
| `ATLASSIAN_MCP_LIST_ARTIFACTS` | Listartifacts | List artifacts created by the current user using cursor pagination. |
| `ATLASSIAN_MCP_LIST_ASSETS_OBJECT_SCHEMAS` | Listassetsobjectschemas | List all Assets object schemas in the selected site's workspace. |
| `ATLASSIAN_MCP_LIST_ASSETS_OBJECT_TYPE_ATTRIBUTES` | Listassetsobjecttypeattributes | List attributes for one or more Assets object types in the selected site. |
| `ATLASSIAN_MCP_LIST_CAPACITY_PLAN_NON_PROJECT_WORK` | Listcapacityplannonprojectwork | Search the non-project work items on a site. "Non-project work" is the user-facing name for a custom contribution target - the buckets of time that are not a Jira issue or an Atlas project, such as Leave, Run the Business (RtB), on-call, or training. Non-project work is an entity that belongs to Capacity Planning: it is defined and stored in Capacity Planning itself, and exists nowhere else in Atlassian. Capacity can be allocated to three kinds of work item - Jira issues, Atlas projects, and non-project work - and only the third is owned by Capacity Planning, so it can only be found through this tool. There is no Jira or Confluence equivalent to look up instead. USE WHEN you need the ARI of a non-project work item before allocating someone to it, or when the user asks what non-project work categories exist. The returned ari is what feeds workId on updateCapacityPlanAllocations and workDataIds on getCapacityPlanAllocations - resolve it here rather than fabricating one. The query is case-insensitive and matches on name; omit it to list everything. Paginate with first/after or last/before. |
| `ATLASSIAN_MCP_LIST_CODE_DIRECTORY` | Listcodedirectory | List the files and subdirectories under a path in a connected source-control repository at a specific commit. Paginated — pass the returned nextPage cursor to fetch the next page. Use it to browse a repository's structure discovered via searchCode. |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_ATTACHMENTS` | Listconfluenceattachments | List attachment metadata for a Confluence content item |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_COMMENTS` | Listconfluencecomments | List comments on a Confluence page or blog post |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_CONTENT` | Listconfluencecontent | List Confluence content of one type within a single space (page, blog post, live doc, whiteboard, database, embed, smart link, folder, slide) |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_CONTENT_VERSIONS` | Listconfluencecontentversions | List version history for a Confluence page, live doc, or blog post, newest first |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_SPACES` | Listconfluencespaces | List Confluence spaces accessible to the caller |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_TASKS` | Listconfluencetasks | List Confluence inline tasks, optionally scoped to a page or blog post and filtered by status |
| `ATLASSIAN_MCP_LIST_CONFLUENCE_TEMPLATES` | Listconfluencetemplates | List Confluence templates and blueprints |
| `ATLASSIAN_MCP_LIST_JIRA_BOARDS` | Listjiraboards | List Jira boards visible to the current user. Use it to get board IDs for getJiraBoardIssueData, getJiraBoardConfig and listJiraBoardSprints; getJiraBoardSprintData can resolve a board by name directly. Returns each board's ID, name, type, edit/privacy/favourite flags and location. Page with startAt until isLast is true. projectKeyOrId matches boards whose JQL references that project, not boards located in it. includePrivate appends private boards, but Jira omits their name and type. |
| `ATLASSIAN_MCP_LIST_JIRA_BOARD_SPRINTS` | Listjiraboardsprints | List a board's sprints — past, active or upcoming — or resolve a sprint name to the numeric sprintId that transitionJiraIssue and manageJiraSprint take. Not needed before createJiraIssue, whose assignToSprint accepts a sprint name or "active" directly. For the active sprint plus its issues and progress in one call, use getJiraBoardSprintData. |
| `ATLASSIAN_MCP_LIST_JIRA_DASHBOARDS` | Listjiradashboards | Returns Jira dashboards visible to the current user for discovery and linking workflows. Use this to find dashboard IDs, names, and links that can be shared or opened. This tool is intentionally lightweight and does not fetch dashboard gadget data. Page with startAt until isLast is true. |
| `ATLASSIAN_MCP_LIST_JIRA_FILTERS` | Listjirafilters | Search saved Jira filters visible to the current user, returning their IDs and JQL. To get the matching issues, run a filter's jql through searchJiraIssuesUsingJql; to build a board, pass its id as createJiraBoard's filterId. projectId finds filters shared with that project, not filters whose JQL mentions it. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_ASSIGNABLE_USERS` | Listjiraissueassignableusers | List the users who can actually be assigned work in a project or on an issue. Filters by assign permission, which a plain name search does not, so prefer it when choosing an assignee for editJiraIssue or createJiraIssue. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_CHANGELOGS` | Listjiraissuechangelogs | Get a Jira issue's field-change history — field edits only — paginated oldest-first with no ordering options. For the current value or most recent change, read total (which counts entries, not items) and request startAt = max(0, total - your page size); for counts or patterns across the history, page from startAt 0. A field's original value is in the first item that changed it; fields never changed leave no trace, and the log starts at the first edit, not at creation. On an item, from/to are IDs or raw values and are absent on text and array fields; fromString/toString are display text, and on array fields like labels each side is the complete set (space-joined, sorted) — diff the two sides, never read one as a delta. Jira omits from and fromString together when the field had no previous value, and to and toString together when it was cleared (so an assignee item with no to/toString is an unassignment). Match items on field, not position: order within an entry varies. Do not infer automation from author.accountType — a script running under a person's token reads "atlassian". |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_COMMENTS` | Listjiraissuecomments | Get a paginated list of comments for a Jira issue. Use this instead of getJiraIssue when an issue has many comments — getJiraIssue embeds at most 20 and cannot page them. Page with startAt until isLast is true; total is the full comment count. Each comment reports any visibility restriction on it and, on JSM, whether it is internal (jsdPublic: false) — check both before quoting or summarising a comment. Comments are returned as a flat list: a non-null parentId means the comment is a reply, and parentId identifies its root/parent comment. Group comments by parentId to reconstruct threads. Set reactionDetail when emoji reaction data is needed. For the issue's own fields, use getJiraIssue. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_LINK_TYPES` | Listjiraissuelinktypes | List the issue link types configured on this site, each with its name and its inward and outward wording (name "Blocks", inward "is blocked by", outward "blocks"). Call it before createJiraIssueLink, which takes the name, when the user has described a link by its inward/outward wording. Issue linking must be enabled site-wide; where it is turned off this operation and createJiraIssueLink both fail. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_REMOTE_ISSUE_LINKS` | Listjiraissueremoteissuelinks | List the Confluence pages and external URLs linked from an issue's Web links section. These are not issue-to-issue links: for those call getJiraIssue with view "evidence", which returns fields.issuelinks. Check here before calling createJiraIssueRemoteIssueLink: the same URL added twice becomes two separate links, and no operation removes one. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_TRANSITIONS` | Listjiraissuetransitions | List the workflow transitions this issue can take right now, for the calling user. Pass a transition's id to transitionJiraIssue. A transition's name and the status it leads to are separate fields: when the user names a target status, match the status a transition leads to, not the transition's name. For every status a project defines, rather than this issue's currently valid moves, use listJiraStatuses. |
| `ATLASSIAN_MCP_LIST_JIRA_ISSUE_WORKLOGS` | Listjiraissueworklogs | Get a paginated list of worklogs (logged time) for a Jira issue. Ordered by worklog creation time, not by when the work started — so the last page is not the most recent work. To find recent work, filter with startedAfter/startedBefore, or read every page and sort by `started`. |
| `ATLASSIAN_MCP_LIST_JIRA_PROJECT_COMPONENTS` | Listjiraprojectcomponents | List a Jira project's components with their ids, names, lead and assignee settings, and issueCount — 0 means no issues use a Jira component; Compass results list only components already linked to issues. realAssignee is who actually receives issues when the configured assignee cannot be used. Use it to find a valid component name before setting components on createJiraIssue or editJiraIssue. |
| `ATLASSIAN_MCP_LIST_JIRA_PROJECT_ISSUE_TYPES_METADATA` | Listjiraprojectissuetypesmetadata | Returns all issue types available in a Jira project, useful for knowing which issue types can be created. Each entry gives the issue type's id, name, description and subtask flag; pass that id as issueTypeId to getJiraIssueTypeMetaWithFields to get the type's create fields. Page with startAt until it covers total. |
| `ATLASSIAN_MCP_LIST_JIRA_PROJECTS` | Listjiraprojects | Find the Jira projects the calling user can see — use it to resolve the projectKey that createJiraIssue requires, or the project a JQL search filters on. For a project's releases use getJiraProjectVersions; for its statuses, use listJiraStatuses. |
| `ATLASSIAN_MCP_LIST_JIRA_STATUSES` | Listjirastatuses | List the statuses a Jira issue can be in — for a project (per work type), for one status looked up by ID or name, or across the whole site by name or category. Takes a PROJECT key (or project ID), not an issue key. For what a specific issue can do right now, use listJiraIssueTransitions instead; a status existing in a project does not mean a given issue can move to it. |
| `ATLASSIAN_MCP_LIST_LOOM_FOLDERS` | Listloomfolders | List the folders in the current Loom workspace that a video can be moved into, returning each folder's id, name, and parent space. Use this to discover a destination folder id before calling moveLoomVideoToFolder, or to pick a folderId for createLoomVideoUpload / publishLoomVideo. Pass query to filter folders by name; omit it to list all folders. |
| `ATLASSIAN_MCP_LIST_LOOM_VIDEOS` | Listloomvideos | List the Loom videos you own (that you created), most recent first. This is a primary entry point for finding a video: it returns each video's ARI, name, and description, which you can pass to getLoomVideo, getLoomVideoDownloadUrl, or any video action. Always scoped to the current user — there is no user parameter. Optionally filter by the video's creation date with the ISO-8601 createdAtFrom / createdAtTo bounds, and paginate with limit / offset. To find videos other people shared with you, use listLoomVideosSharedWithMe. |
| `ATLASSIAN_MCP_LIST_LOOM_VIDEOS_SHARED_WITH_ME` | Listloomvideossharedwithme | List the Loom videos that other people have shared with you, most recent first. Returns each video's ARI, name, and description so you can pass the ARI to getLoomVideo, getLoomVideoDownloadUrl, or any video action. Always scoped to the current user — there is no user parameter. Optionally filter by the video's creation date with the ISO-8601 createdAtFrom / createdAtTo bounds (this filters on when the video was created, not when it was shared with you) and paginate with limit / offset. To list videos you own, use listLoomVideos. |
| `ATLASSIAN_MCP_LOOKUP_JIRA_ACCOUNT_ID` | Lookupjiraaccountid | Look up a Jira user account ID by display name or email — skip it if you already have the accountId. Returns only accountId, displayName and active, plus emailAddress when profile visibility allows it; for time zone, locale, groups, or application roles, pass the accountId on to getJiraUser. To choose an assignee use listJiraIssueAssignableUsers instead, which returns only users who can be assigned. |
| `ATLASSIAN_MCP_MANAGE_CONFLUENCE_APPROVAL` | Manageconfluenceapproval | Request, decide, cancel, clear, or update a Confluence page approval. Approvals support pages only, not live docs or blog posts. |
| `ATLASSIAN_MCP_MANAGE_JIRA_PROJECT_VERSION` | Managejiraprojectversion | Create, update, release, or archive a Jira project version (release or fix version). create returns the version id the other actions need. If release uses moveUnfixedIssuesTo, setting released: false later does not move those issues back. Otherwise release and archive change only the version, not its issues, and update can reverse either. To view or list versions, or read their release driver and approvers, use getJiraProjectVersions. |
| `ATLASSIAN_MCP_MANAGE_JIRA_PROJECT_VERSION_RELATED_WORK` | Managejiraprojectversionrelatedwork | Create or update related work links on a Jira project version (release). To read a version's existing related work, use getJiraProjectVersionRelatedWork. Both actions return the relatedWorkId a later update requires, with the stored category, title and url. |
| `ATLASSIAN_MCP_MANAGE_JIRA_SPRINT` | Managejirasprint | Create, add issues to, start, update, or close a Jira sprint; get a sprintId from listJiraBoardSprints. create returns the new sprint's id, which every later action needs. update rewrites the whole sprint, so any of name, goal, startDate or endDate you omit is cleared — resend the ones to keep; start and close keep them. Closing is one-way. addIssues returns movedIssues, as does start when given issueIdsOrKeys; both add estimatedIssues, with the field each estimate was written to, when storyPoints are passed. |
| `ATLASSIAN_MCP_MOVE_CONFLUENCE_CONTENT` | Moveconfluencecontent | Move a Confluence page, whiteboard, database, folder, or embed to a new parent or sibling position |
| `ATLASSIAN_MCP_MOVE_LOOM_VIDEO_TO_FOLDER` | Moveloomvideotofolder | Move a Loom video into a folder in the current workspace. Get a destination folder id from listLoomFolders (or createLoomFolder). Identify the video by ARI or share URL. |
| `ATLASSIAN_MCP_PUBLISH_ARTIFACT` | Publishartifact | Step 2 of 2 for uploading an artifact. Finalizes content whose bytes were already transferred via createArtifactsUpload (step 1), using the contentId from that step's completionPayload. Only call after the upload has completed. Creates a new artifact, or publishes a new version by supplying an existing artifact ID. Save, preserve, or store AI-generated analysis, reports, dashboards, charts, visualizations, and HTML slide decks durably behind a persistent link. Returns that link, which does not by itself grant anyone access: use shareArtifact to let named users, groups, or teams open it. |
| `ATLASSIAN_MCP_PUBLISH_LOOM_VIDEO` | Publishloomvideo | Step 2 of 2 for uploading a new video to Loom. Finalizes a video whose bytes were already transferred via createLoomVideoUpload (step 1), using the keyPath and fileName from that step's completionPayload. Only call this after the upload has completed. Returns the user-facing Loom share URL. Optionally place the video in a folder or space. |
| `ATLASSIAN_MCP_QUERY_ASSETS_OBJECTS` | Queryassetsobjects | Query Assets objects with generated AQL for one object type in the selected site. |
| `ATLASSIAN_MCP_QUERY_ASSETS_OBJECT_SCHEMA_ATTRIBUTES` | Queryassetsobjectschemaattributes | Query attributes for one or more Assets object schemas in the selected site. |
| `ATLASSIAN_MCP_QUERY_ASSETS_OBJECTS_USING_AQL` | Queryassetsobjectsusingaql | Query Assets objects with raw or generated AQL in the selected site. |
| `ATLASSIAN_MCP_QUERY_ASSETS_OBJECT_TYPES` | Queryassetsobjecttypes | Query Assets object types by schema in the selected site. |
| `ATLASSIAN_MCP_QUERY_ASSETS_REFERENCE_TYPES` | Queryassetsreferencetypes | Query global or schema-scoped Assets reference types in the selected site. |
| `ATLASSIAN_MCP_RECOVER_LOOM_VIDEO` | Recoverloomvideo | Recover a Loom video that failed to process, by initiating a recovery job to re-process it. Use this when a video is stuck or errored (for example isComplete is false in getLoomVideo). Identify the video by ARI or share URL. |
| `ATLASSIAN_MCP_REMOVE_CONFLUENCE_CONTENT_PERMISSIONS` | Removeconfluencecontentpermissions | Removes content-level permission grants from a single Confluence content entity. Optionally narrow what is removed by operation, principal-type, and/or principal-id. Omitting all three filters clears ALL Specific access grants on the content. This does not change General access. This can revoke access. The server protects against the current user locking themselves out of their own content. |
| `ATLASSIAN_MCP_REOPEN_CONFLUENCE_TASK` | Reopenconfluencetask | Reopen a completed Confluence inline task (mark it incomplete) |
| `ATLASSIAN_MCP_REPLACE_CONFLUENCE_CONTENT_PERMISSIONS` | Replaceconfluencecontentpermissions | Replaces ALL content-level permission grants on a single Confluence content entity with the supplied set. Grants not present in the request are removed, so this can revoke access — prefer addConfluenceContentPermissions when you only intend to add grants. Pass the full desired grant set via the `permissions` input, an array of objects shaped exactly like: [ { "operation": "update", "principals": [ { "principalType": "user", "principalId": "" }, { "principalType": "group", "principalId": "" } ] } ] Enums: operation is one of read \| update; principalType is one of user \| group. For a user, principalId is the Atlassian account ID; for a group, principalId is the group ID. operation `read` = can view, `update` = can edit (edit includes view). A principal has only ONE effective level, so listing a principal under `read` when they should keep editing leaves them view-only — list them under `update` to preserve edit. Never list the same principal under both operations. Replacing Specific access does not change General access. Because this REPLACES the whole set, always include yourself with `update` if the content is (or is becoming) EDIT_RESTRICTED / VIEW_RESTRICTED. The server protects against locking the current user out of their own content. |
| `ATLASSIAN_MCP_RESOLVE_CONFLUENCE_CONTENT_MACROS` | Resolveconfluencecontentmacros | Resolve supported Confluence macro wrappers into per-macro HTML displayed in Confluence web. Use after getConfluenceContent when a full document read returns wrappers instead of visible content, especially for Include Page, Excerpt Include, and Jira macros. Use the returned HTML bodies to supplement the base document; this operation does not reconstruct the full page. |
| `ATLASSIAN_MCP_RESTORE_CONFLUENCE_CONTENT_VERSION` | Restoreconfluencecontentversion | Restore an earlier version of a Confluence page, live doc, or blog post as a new latest version |
| `ATLASSIAN_MCP_REVOKE_ARTIFACT_SHARE` | Revokeartifactshare | Stop sharing an artifact with named users, groups, or teams, or with every audience at once. Supply either audienceAris or revokeAll, never both and never neither. Revoking a share that was never granted still succeeds. This does not change the artifact's access level. An artifact readable by anyone with the link, or by the whole organisation, stays readable after every share is revoked, and the response says so when that is the case. |
| `ATLASSIAN_MCP_SCAN_CODE_REPO` | Scancoderepo | List the file paths in a connected source-control repository at a specific commit, optionally filtered by a glob or restricted to a subtree. Returns paths only, no file contents. Paginated — pass the returned nextCursor to fetch the next page. |
| `ATLASSIAN_MCP_SEARCH` | Search | Semantically search Jira, Confluence, and connected third-party content using Rovo Search. ALWAYS use this tool to search for Jira and Confluence content unless CQL or JQL is specified. The utilisation of this tool may consume up to 10 Rovo Credits. If the user or organization does not have any Rovo Credits available, the tool call will fail. |
| `ATLASSIAN_MCP_SEARCH_ASSETS_OBJECTS` | Searchassetsobjects | Search Assets objects by name or query-builder filters in the selected site. |
| `ATLASSIAN_MCP_SEARCH_CODE` | Searchcode | Search source code across connected source-control providers using semantic, lexical, or hybrid matching. Returns matching files along with the specific code spans that matched. |
| `ATLASSIAN_MCP_SEARCH_CONFLUENCE` | Searchconfluence | Search Confluence content with CQL |
| `ATLASSIAN_MCP_SEARCH_FOCUS_AREAS` | Searchfocusareas | Search focus areas using text and structured filters with optional pagination and sorting. Omitting query and filters lists/browses focus areas. Exact lookup by ARI, UUID, or URL should use getFocusArea instead. getFocusArea also returns more detail per focus area than search results do — startDate, archived/draft/starred/watching flags, created and updated timestamps, and direct child focus areas — so follow up with it once a specific focus area is identified. Set include.aboutContent to true to fetch about content, and/or pass include.statusUpdates for latest status updates per result. Paginate with first/after using pageInfo.endCursor. Custom field filtering is supported through filters.customFields (resolve field searchKeys and select option UUIDs with getFocusAreaCustomFields first); filtering for focus areas where a custom field has no value is not supported — do not retry with other operators to express it. |
| `ATLASSIAN_MCP_SEARCH_GOALS` | Searchgoals | Find Atlassian Goals by text and structured filters. |
| `ATLASSIAN_MCP_SEARCH_JIRA_ISSUES_USING_JQL` | Searchjiraissuesusingjql | Search for Jira issues using JQL, with an optional approximate count. Use view: "evidence" to fetch and map custom fields. Page with nextPageToken until isLast is true. For one known issue, use getJiraIssue. JQL only returns issues; inferring anything else from issue rows is silently incomplete. Use the purpose-built operation for sprints, boards, board config, field history, worklogs, releases, accountIds, issue types, projects and statuses. |
| `ATLASSIAN_MCP_SEARCH_JSM_REQUESTS` | Searchjsmrequests | Search one bounded page of caller-visible Jira Service Management requests. Use JQL for project, request type, assignee, status, approval, text, due-date, and ordering filters. Use this tool when JSM requests, issues, or work items need your attention, when you need urgency signals from SLA state or sentiment, or when you need to identify auto-resolvable work items after approval. Returns summary, status, assignee, request type, sentiment, SLA state, card display data (issue-type icon, status category, assignee and requester avatars), URLs, and compact resolver-plan availability. This tool does not return full request details or plan steps. When the upstream request search ranks rows, it preserves the total score, each score component, and evidence so clients can render and explain urgency consistently. |
| `ATLASSIAN_MCP_SEARCH_PROJECTS` | Searchprojects | Find Atlassian Projects by text and structured filters. |
| `ATLASSIAN_MCP_SEARCH_TALENT_POSITIONS` | Searchtalentpositions | Search Talent positions by their own attributes using an RQL filter, returning a paginated list of positions with their requested field values (each position's ARI is the `id`, so this is how you obtain ARIs for getTalentPosition). Use this for flat attribute filters — job title, country, level, status, job family, tenure. Use getTalentPositionsByEntity INSTEAD when scoping to a reporting line or a focus area, and getTalentGroupMetrics INSTEAD when you only need counts rather than the positions themselves. RQL field IDs and their valid values are workspace-specific and must not be guessed — call getTalentFields to resolve them. The common built-ins are positionJobTitle, positionCountry, positionLevel, positionStatus, positionRole, positionManager, positionTenure, positionJobFamily, positionEmploymentType, positionWorker and team; for anything else, or for a field's valid values, go through getTalentFields. If a filter returns nothing, suspect the field ID or its value and check it with getTalentFields rather than adding more filters. |
| `ATLASSIAN_MCP_SEARCH_TEAMS` | Searchteams | Search Atlassian teams by name with cursor pagination. Core team fields are returned by default; pass expand ["members"] to include a separately paginated member summary for each team. |
| `ATLASSIAN_MCP_SET_CONFLUENCE_CONTENT_RESTRICTION_STATE` | Setconfluencecontentrestrictionstate | Sets the high-level restriction state for a single Confluence content entity. Provide `state` as one of: OPEN — "Open, Anyone in this space can edit": everyone with space access can VIEW and EDIT. EDIT_RESTRICTED — "Open, Anyone in this space can view": everyone with space access can VIEW; only Specific access update principals can edit. VIEW_RESTRICTED — "Restricted, Only specific people can view or edit": only Specific access read/update principals can VIEW; only update principals can edit. HOW GENERAL AND SPECIFIC ACCESS WORK TOGETHER: the state is the General access baseline. Moving away from OPEN removes broad access. Permission grants are Specific access exceptions that add access for individual users or groups; they do not remove General access or make content private by themselves. A principal has exactly ONE Specific access level — update, read, or none — and update includes read/view. Changing the state does not create or modify Specific access grants. SELF-LOCKOUT: when leaving OPEN, the caller must retain `update`; `read` is not sufficient for editing. Private creation already performs the safe VIEW_RESTRICTED + creator update setup. Existing-state and self-lockout validation belongs to the Confluence monolith, and this tool never disables it. |
| `ATLASSIAN_MCP_SET_CONFLUENCE_CONTENT_STATUS` | Setconfluencecontentstatus | Set the content state (status) of a Confluence page or blog post |
| `ATLASSIAN_MCP_SET_CONFLUENCE_SPACE_INSTRUCTIONS` | Setconfluencespaceinstructions | Set the space instructions for a Confluence space |
| `ATLASSIAN_MCP_SET_JSM_AGENT_AVAILABILITY` | Setjsmagentavailability | Set one Jira Service Management Workforce Management agent's availability for a service project. Use this for "set me available", "set me on break", or an explicitly supplied agentAccountId when you are authorized to update that agent. Status accepts AVAILABLE or ON_BREAK and TWG-compatible aliases available, active, online, away, on-break, or unavailable. This is not for reading availability, bulk team updates, project defaults, assignment, or Live Chat status. |
| `ATLASSIAN_MCP_SHARE_ARTIFACT` | Shareartifact | Grant named users, groups, or teams permission to view an existing artifact. Use this when someone cannot open an artifact yet, because publishArtifact returns a link but grants nobody access, and this is what makes that link work for them. This adds individual grants; it does not make the artifact readable by anyone with the link or by the whole organisation, which are the artifact's access level and are not set here. The call is atomic, so if any audience is invalid nothing is granted. Returns the artifact's resulting access level and the audiences it is now shared with. |
| `ATLASSIAN_MCP_SHARE_LOOM_VIDEO` | Shareloomvideo | Share a Loom video with one or more specific Atlassian users by sending them a notification (with an optional message). This notifies people; it does not change the video's visibility — to make a video WORKSPACE- or PUBLIC-visible use updateLoomVideoPermissions instead. Identify the video by ARI or share URL, and recipients by their Atlassian user ARIs or exact name/email address. |
| `ATLASSIAN_MCP_STAR_CONFLUENCE_CONTENT` | Starconfluencecontent | Star a Confluence content item for the current user |
| `ATLASSIAN_MCP_STAR_CONFLUENCE_SPACE` | Starconfluencespace | Star a Confluence space for the current user |
| `ATLASSIAN_MCP_TRANSITION_JIRA_ISSUE` | Transitionjiraissue | Move a Jira issue through its workflow and/or into a sprint or the backlog — at least one is required. Take transitionId from listJiraIssueTransitions; transitionName is the transition's own name, not the target status. Returns the status the issue actually landed in, so no read-back is needed. If the sprint/backlog move fails after the status change succeeds, the status change stays — retry the move alone, not the whole call. |
| `ATLASSIAN_MCP_TRANSITION_JSM_REQUEST` | Transitionjsmrequest | Perform one customer/request-portal-visible Jira Service Management request transition selected from getJsmRequestActionMetadata. This is for help-seeker/customer portal transitions, not agent-side queue or generic Jira workflow transitions. It returns compact execution metadata only; use getJsmRequest afterwards when refreshed request state is needed. |
| `ATLASSIAN_MCP_UNARCHIVE_CONFLUENCE_CONTENT` | Unarchiveconfluencecontent | Unarchive (restore) a single Confluence content entity |
| `ATLASSIAN_MCP_UNSTAR_CONFLUENCE_CONTENT` | Unstarconfluencecontent | Remove a star from a Confluence content item for the current user |
| `ATLASSIAN_MCP_UNSTAR_CONFLUENCE_SPACE` | Unstarconfluencespace | Remove a star from a Confluence space for the current user |
| `ATLASSIAN_MCP_UNWATCH_CONFLUENCE_CONTENT` | Unwatchconfluencecontent | Stop watching a Confluence content item for the current user |
| `ATLASSIAN_MCP_UNWATCH_CONFLUENCE_LABEL` | Unwatchconfluencelabel | Stop watching a Confluence label for the current user |
| `ATLASSIAN_MCP_UNWATCH_CONFLUENCE_SPACE` | Unwatchconfluencespace | Stop watching a Confluence space for the current user |
| `ATLASSIAN_MCP_UPDATE_CAPACITY_PLAN_ALLOCATIONS` | Updatecapacityplanallocations | Allocate people to a single work item over one or more time periods in Individual Capacity Planning. USE WHEN the user asks to put someone on a work item for a period, to change how much of their time it takes, to book leave, or to clear an allocation. Every person in contributorDataIds is assigned to the work item automatically, so it does not need to be assigned to them first, and every allocation is applied to every person - the number of writes is contributorDataIds x allocations, which must not exceed 100. Only one work item per call. Resolve a non-project work ARI with listCapacityPlanNonProjectWork rather than fabricating one. Each allocation covers a period and carries the TOTAL effort for that period, of any length. To express a RATE, send one allocation per interval: "2 days a week for three weeks" is three allocations of 2, NOT one allocation of 2 spanning three weeks. A single allocation spanning all of September means "this much effort across September", which the plan spreads over the weeks in it. Re-running the same periods REPLACES those allocations rather than adding to them, so the call is safe to repeat; set value to 0 to clear a period. Because value is the TOTAL for the period and re-running replaces it, any RELATIVE change - "add a day", "bump Chris to 3 days", "halve it" - needs the current figure first. Call getCapacityPlanAllocations with the same contributorDataIds and workDataIds, the same date window, and expand: ["contributions"], then send the new total. Writing the relative amount directly overwrites the existing allocation instead of adding to it. An absolute request ("put Chris on PROJ-42 at 50% in September") needs no prior read. This does NOT add anyone to a capacity plan and cannot create work items - contributors and work items are global, so an allocation written here shows up on every plan they belong to, matching the UI. Add 'contributions' to expand to return the value written per period and confirm what landed, 'contributorAggregations' for each person's total contribution and availability across ALL their work in the allocated window (use this to report whether the change has pushed someone over capacity), and 'workAggregations' for the work item's total effort across ALL contributors. Both aggregations come back as separate top-level objects. Omit expand when a simple confirmation is enough - each value costs extra work upstream. Capacity plans assume a baseline 40-hour working week, so 5 DAYS, 40 HOURS, and 100 PERCENTAGE are all the same amount - say so when reporting days or percentages. |
| `ATLASSIAN_MCP_UPDATE_CAPACITY_PLAN_NON_PROJECT_WORK` | Updatecapacityplannonprojectwork | Rename an existing non-project work item in Individual Capacity Planning. "Non-project work" is a bucket of time that is not a Jira issue or an Atlas project, such as Leave, Run the Business (RtB), on-call, or training. USE WHEN the user asks to rename an existing non-project work item. This only changes the item's name; it does not move allocations, add contributors, or delete anything. To change who is allocated or how much effort a period carries, use updateCapacityPlanAllocations instead. Resolve the item's id first with listCapacityPlanNonProjectWork rather than fabricating one. Returns whether the update succeeded and any errors. |
| `ATLASSIAN_MCP_UPDATE_CONFLUENCE_COMMENT` | Updateconfluencecomment | Update an existing Confluence footer or inline comment on a page or blog post |
| `ATLASSIAN_MCP_UPDATE_CONFLUENCE_COMMENT_RESOLUTION` | Updateconfluencecommentresolution | Resolve or reopen a Confluence footer or inline comment |
| `ATLASSIAN_MCP_UPDATE_CONFLUENCE_CONTENT` | Updateconfluencecontent | Update Confluence docs, whiteboards, or databases. Before choosing body.format or authoring an edit, call getContentFormatGuide through execute with name=getContentFormatGuide and inputs={toolName}. Use updateConfluencePage for pages, blog posts, and live docs; updateConfluenceWhiteboard for whiteboards; or updateConfluenceDatabase for databases. Space instructions describe how to author content in a Confluence space. Apply relevant guidance to the content you create or update. Follow this decision rule: 1. Before authoring, call getConfluenceSpace for the target space unless getConfluenceContent returned metadata.hasSpaceInstructions=false for content in that space. If getConfluenceSpace already returned for the same cloudId and space during the current task, reuse its spaceInstructions and do not call it again. 2. When the value is true or unknown and instructions have not been loaded, you must call getConfluenceSpace once before authoring through execute with name=getConfluenceSpace, top-level cloudId, and inputs={spaceIdOrKey}. 3. Apply any returned spaceInstructions. If a successful response omits them, no instructions are configured; author with defaults and do not block. After a successful getConfluenceSpace call, do not make a separate getConfluenceSpaceInstructions call. |
| `ATLASSIAN_MCP_UPDATE_FOCUS_AREA` | Updatefocusarea | Update one or more fields on an existing Focus area. Provide at least one of name, owner, startDate, or description — identifier-only calls fail upstream. Omitted fields are unchanged. Use createFocusAreaStatusUpdate for status, health, target date, or status summary changes. Multiple fields become separate mutations — a later failure can leave earlier fields already applied. |
| `ATLASSIAN_MCP_UPDATE_JSM_OPS_ALERT` | Updatejsmopsalert | Update one Jira Service Management Operations alert by acknowledging, unacknowledging, closing, or escalating it. Escalation requires an escalation responder ID assigned to that alert. Optionally poll the asynchronous request status once. |
| `ATLASSIAN_MCP_UPDATE_LOOM_VIDEO_PERMISSIONS` | Updateloomvideopermissions | Change who can access a Loom video by setting its visibility to OWNER (only you), WORKSPACE (anyone in your workspace), or PUBLIC (anyone with the link). The change is immediate and permanent. To notify specific people about a video instead of changing its visibility, use shareLoomVideo. Identify the video by ARI or share URL. Returns `id`, `name`, `visibility`, `url`, and `updatedAt`. Omit responseFields to keep these defaults; `videoUrl` and `videoAri` are inputs, not response field names. |
| `ATLASSIAN_MCP_UPDATE_LOOM_VIDEO_TITLE` | Updateloomvideotitle | Rename a Loom video. The change is immediate and visible to all viewers. Identify the video by ARI or share URL. To change who can access the video instead of its title, use updateLoomVideoPermissions. |
| `ATLASSIAN_MCP_UPDATE_TALENT_POSITION_FOCUS_AREA_MAPPINGS` | Updatetalentpositionfocusareamappings | Allocate Talent positions to focus areas, or clear the focus area a position is allocated to. Each mapping sets the focus area of ONE position and REPLACES the focus area that position is allocated to now — a position is allocated to one focus area at a time, so this is not additive and it cannot add a second focus area to a position. Omit focusAreaId in a mapping to clear the allocation and leave that position unallocated. Every mapping is applied as a single all-or-nothing change, so one rejected ARI fails the whole call — keep batches small. Do not guess either ARI: take a position ARI from a searchTalentPositions result's id or a getTalentPositionsByEntity result's entityId, and take a focus area ARI from getTalentFields for the focus area field, or from searchFocusAreas. To see which positions a focus area is staffed by before or after the change, call getTalentPositionsByEntity with entity=focusArea. |
| `ATLASSIAN_MCP_UPDATE_TEAM` | Updateteam | Update one Atlassian team through exactly one operation. Use update_details with displayName, description, or membershipSettings; add_members with accountIds and/or add-only emails; or remove_members with accountIds. Metadata and membership changes cannot be mixed. Removing members can remove access. |
| `ATLASSIAN_MCP_UPLOAD_ATTACHMENT_TO_JIRA_ISSUE` | Uploadattachmenttojiraissue | Attach a single local file to a Jira issue. Two phases: (1) call with filePath only to get a local uploadCommand; run it from your shell to upload the bytes and obtain a fileId; (2) call again with that fileId to attach the file to the issue. To instead embed the uploaded file inline in a body, stop after phase 1 and pass the fileId (with its collection) to editJiraIssue/addOrEditJiraIssueComment via inlineFileId (appends at the end) or an HTML body media node (for positional placement) — do not also run phase 2 for the same file. |
| `ATLASSIAN_MCP_WATCH_CONFLUENCE_CONTENT` | Watchconfluencecontent | Start watching a Confluence content item for the current user |
| `ATLASSIAN_MCP_WATCH_CONFLUENCE_LABEL` | Watchconfluencelabel | Start watching a Confluence label for the current user |
| `ATLASSIAN_MCP_WATCH_CONFLUENCE_SPACE` | Watchconfluencespace | Start watching a Confluence space for the current user |
| `ATLASSIAN_MCP_WATCH_JIRA_ISSUE` | Watchjiraissue | Add or remove the calling user as a watcher on a Jira issue. Always acts on you — it cannot add or remove another person, and it does not list who is watching. Watching only subscribes you to notifications about the issue; it does not assign it to you. To do that, set the issue's assignee field with editJiraIssue. |

## Supported Triggers

None listed.

## Creating MCP Server - Stand-alone vs Composio SDK

Once connected, your Grok bot can access the Atlassian MCP server through Composio to run the actions you authorize, using plain language in the chat.

## Complete Code

None listed.

## Conclusion

### Way Forward
Now that Atlassian is connected, extend your setup by connecting the other apps you already use every day, so your Grok bot can run true cross-app workflows end to end.
- Connect Calendar to turn threads into scheduled meetings automatically.
- Connect Slack or Teams to post summaries, approvals, and alerts where your team works.
- Connect Notion, Linear, Jira, or Asana to convert requests into tickets, tasks, and docs.
- Connect Drive, Dropbox, or OneDrive to fetch, file, and share attachments without manual steps.
Start with one workflow you do repeatedly, then keep adding apps as you find new handoffs. With everything behind a single connection, your bot can coordinate multiple tools safely and reliably in one conversation.

## How to build Atlassian MCP Agent with another framework

- [ChatGPT](https://composio.dev/toolkits/atlassian_mcp/framework/chatgpt)
- [Claude Cowork](https://composio.dev/toolkits/atlassian_mcp/framework/claude-cowork)
- [Hermes](https://composio.dev/toolkits/atlassian_mcp/framework/hermes-agent)
- [Atomic Agent](https://composio.dev/toolkits/atlassian_mcp/framework/atomic-agent)

## Related Toolkits

- [Google Sheets](https://composio.dev/toolkits/googlesheets) - Google Sheets is a cloud-based spreadsheet tool for real-time collaboration and data analysis. It lets teams work together from anywhere, updating information instantly.
- [Notion](https://composio.dev/toolkits/notion) - Notion is a collaborative workspace for notes, docs, wikis, and tasks. It streamlines team knowledge, project tracking, and workflow customization in one place.
- [Airtable](https://composio.dev/toolkits/airtable) - Airtable combines the flexibility of spreadsheets with the power of a database for easy project and data management. Teams use Airtable to organize, track, and collaborate with custom views and automations.
- [Asana](https://composio.dev/toolkits/asana) - Asana is a collaborative work management platform for teams to organize and track projects. It streamlines teamwork, boosts productivity, and keeps everyone aligned on goals.
- [Google Tasks](https://composio.dev/toolkits/googletasks) - Google Tasks is a to-do list and task management tool integrated into Gmail and Google Calendar. It helps you organize, track, and complete tasks across your Google ecosystem.
- [Linear](https://composio.dev/toolkits/linear) - Linear is a modern issue tracking and project planning tool for fast-moving teams. It helps streamline workflows, organize projects, and boost productivity.
- [Jira](https://composio.dev/toolkits/jira) - Jira is Atlassian’s platform for bug tracking, issue tracking, and agile project management. It helps teams organize work, prioritize tasks, and deliver projects efficiently.
- [Clickup](https://composio.dev/toolkits/clickup) - ClickUp is an all-in-one productivity platform for managing tasks, docs, goals, and team collaboration. It streamlines project workflows so teams can work smarter and stay organized in one place.
- [Monday](https://composio.dev/toolkits/monday) - Monday.com is a customizable work management platform for project planning and collaboration. It helps teams organize tasks, automate workflows, and track progress in real time.
- [Addressfinder](https://composio.dev/toolkits/addressfinder) - Addressfinder is a data quality platform for verifying addresses, emails, and phone numbers. It helps you ensure accurate customer and contact data every time.
- [Agiled](https://composio.dev/toolkits/agiled) - Agiled is an all-in-one business management platform for CRM, projects, and finance. It helps you streamline workflows, consolidate client data, and manage business processes in one place.
- [Aha!](https://composio.dev/toolkits/aha) - Aha! is a product management platform for setting strategy, prioritizing ideas, and planning roadmaps. It helps product teams align goals with development work and make better decisions.
- [AirOps MCP](https://composio.dev/toolkits/airops_mcp) - AirOps MCP is AirOps' managed content platform exposing brand context, knowledge bases, and workflows. Use it to give agents brand-aware content, automate content workflows, and track Brand Kit edits.
- [Amazing Marvin](https://composio.dev/toolkits/amazing_marvin) - Amazing Marvin is a personal productivity app for tasks, projects, habits, planning, time tracking, and rewards. It helps you build a flexible workflow that matches how you actually work.
- [Ambiguous MCP](https://composio.dev/toolkits/ambiguous_mcp) - Ambiguous MCP is a provider-hosted workspace service for documents, spreadsheets, tasks, CRM, calendar, mail, chat, and administration. It provides one workspace-scoped connection for managing these resources through natural language.
- [AnyDB](https://composio.dev/toolkits/any_db) - AnyDB is a flexible collaborative database platform for organizing records, files, and automated workflows. It helps teams centralize data, collaborate on records, and automate routine tasks.
- [ApptiveGrid](https://composio.dev/toolkits/apptivegrid) - ApptiveGrid is a collaborative no-code database and app platform for spaces, grids, forms, and structured records. It helps teams organize operational data, collect inputs, and build lightweight business apps without custom backend work.
- [Ascora](https://composio.dev/toolkits/ascora) - Ascora is a cloud-based field service management platform for service businesses. It streamlines scheduling, invoicing, and customer operations in one place.
- [Basecamp](https://composio.dev/toolkits/basecamp) - Basecamp is a project management and team collaboration tool by 37signals. It helps teams organize tasks, share files, and communicate efficiently in one place.
- [Beeminder](https://composio.dev/toolkits/beeminder) - Beeminder is an online goal-tracking platform that uses monetary pledges to keep you motivated. Stay accountable and hit your targets with real financial incentives.

## Frequently Asked Questions

### What are the differences in Tool Router MCP and Atlassian MCP?

With a standalone Atlassian MCP server, the agents and LLMs can only access a fixed set of Atlassian tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Atlassian and many other apps based on the task at hand, all through a single MCP endpoint.

### Can I use Tool Router MCP with Grok Bot?

Yes, you can. Grok Bot fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Atlassian tools.

### Can I manage the permissions and scopes for Atlassian while using Tool Router?

Yes, absolutely. You can configure which Atlassian scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

### How safe is my data with Composio Tool Router?

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Atlassian data and credentials are handled as safely as possible.

---
[See all toolkits](https://composio.dev/toolkits) · [Composio docs](https://docs.composio.dev/llms.txt)
