Search, create, and retrieve tasks, add comments, and track time through natural language commands.
ClickUp MCP demonstrates good overall quality with well-structured tool definitions, comprehensive parameter descriptions, and clear organizational patterns. All 14 tools have descriptions and input schemas with type information. Tool naming follows verb_noun conventions (readDocument, updateTask, searchSpaces). However, there are notable gaps: output schemas are not formally documented for any tool (critical for agent chaining), error handling lacks recovery guidance, some parameter descriptions could be more precise about formats and constraints, and no explicit pagination documentation despite dealing with large result sets.
Adds a comment to a specific task, or a threaded reply to an existing comment when `parent_comment_id` is set. LINKING BEST PRACTICES: - Always reference related tasks using ClickUp URLs (https://app.clickup.com/t/TASK_ID) - Task URLs become live task references (chip with task name and status), so write them bare - any custom link text on a task URL is replaced by the live task name - Include task links when mentioning dependencies, related work, or follow-ups - Link to relevant lists, spaces, or other ClickUp entities when applicable PROGRESS UPDATES: Include current status, progress information, and next steps. FORMATTING: Headings, **bold**, *italic*, ~~strikethrough~~, `inline code`, code blocks, links, blockquotes, bullet/numbered/nested lists and checkboxes (- [ ] / - [x]) all render natively. TABLES ARE NOT SUPPORTED by ClickUp comments - a markdown table is automatically converted to a monospace code block, which is readable but plain. Prefer bold labels or lists over tables when writing comments. IMAGES: Reference images with normal markdown - ``. This server runs locally, so a local file path is read and uploaded automatically - never inline a screenshot as base64 when a path exists, it costs orders of magnitude more tokens. Also accepted: `data:` URIs, http(s) URLs (downloaded and re-uploaded), and existing ClickUp attachment URLs (embedded as-is). The caption becomes the attachment filename, which is what ClickUp displays beneath the image - so write a caption that reads well. IMAGE LAYOUT: An image inside a numbered list breaks ClickUp's numbering. Write walkthrough steps as bold lines with a blank line before and after the image instead (`**1. Open the login page**`). If external links are provided, verify they are publicly accessible and incorporate relevant information. Check the task's current status - if it's in 'backlog' or similar inactive states, suggest moving it to an active status like 'in progress' when work is being done.
Creates a new document or page within an existing document.
Output schemas not documented for any tool. LLMs cannot plan chaining between tools without knowing what fields are returned (e.g., does searchTasks return task_id or id? Is it in a tasks array or flat?). This forces agents to guess field names for downstream tool calls.
Error handling provides no recovery guidance. Tools throw generic fetch errors (e.g., 'Error fetching document: 404 Not Found') without suggesting what the agent should try next. Should categorize as retryable vs user-fixable and offer alternatives.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | C | 65 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 47 | - | v1 |
Creates a new task in a ClickUp list.
Deletes a comment from a task.
Edits an existing comment on a task.
Gets comprehensive information about a list including description and available statuses. ALWAYS use the list URL (https://app.clickup.com/v/l/LIST_ID) when referencing lists. Use this before creating tasks to understand the list context and available statuses for new tasks. IMPORTANT: The list description often contains valuable project context, requirements, or guidelines - read and consider this information when creating or updating tasks in this list. Share the clickable list URL when suggesting list-related actions.
Get a ClickUp task with images and comments by ID. Always use this URL when referencing tasks in conversations or sharing with others. The response provides complete context including task details, comments, and status history.
Get a ClickUp document with page structure and content. Documents can be discovered via searchSpaces (which includes documents in space tree) or by direct URL from the user or within tasks. Always use the document URL when referencing documents in conversations or sharing with others. The response provides complete document metadata, page structure, and requested page content. Document URLs look like this: https://app.clickup.com/docs/DOC_ID/PAGE_ID
Searches spaces (sometimes called projects) by name or ID with fuzzy matching. If 5 or fewer spaces match, automatically fetches all lists (sometimes called boards) and folders within those spaces to provide a complete tree structure. If more than 5 spaces match, returns only space information with guidance to search more precisely. You can search by space name (fuzzy matching) or provide an exact space ID. Always reference spaces by their URLs when discussing projects or suggesting actions.
Searches tasks (sometimes called Tickets or Cards) by name, content, assignees, and ID with fuzzy matching and support for multiple search terms (OR logic). Can filter by multiple list_ids, space_ids, todo status, or tasks assigned to the current user. If no search terms provided, returns most recently updated tasks. Can also be used to find tasks for the current user by providing the assigned_to_me flag. For optimal results, as your ClickUp tasks may be primarily in 'en', consider providing search terms in English and 'en'. Always reference tasks by their URLs when discussing search results or suggesting actions. You'll get a rough overview of the tasks that match the search terms, sorted by relevance. Always use getTaskById to get more specific information if a task is relevant, and always share the task URL.
Adds or updates a time entry for a task.
Updates an existing document page's content and/or name. Use this when you have both doc_id and page_id from readDocument. Content is in markdown format and can be replaced or appended. Always reference documents by their URLs when sharing with users.
Appends documentation or context to a list's description. ALWAYS reference the list URL (https://app.clickup.com/v/l/LIST_ID) when updating or discussing lists. SAFETY FEATURE: Description updates are APPEND-ONLY to prevent data loss - existing content is preserved. Use this to add project context, requirements, or guidelines that LLMs should consider when working with tasks in this list. Include links to related tasks, spaces, or external resources in the appended content. Content is appended in markdown format with timestamp for tracking changes.
Updates an existing ClickUp task with new values for name, description, status, priority, dates, time estimate, tags, or assignees.
Pagination not documented or supported. searchTasks and searchSpaces can return large result sets but no limit, page, offset, or cursor parameters are present. Response structure for pagination (e.g., total count, next_cursor) is undocumented. This risks context window exhaustion with 100+ results.
Parameter format/constraint descriptions are vague. 'due_date as ISO date string (e.g., 2024-10-06T23:59:59+02:00)' mixes format hint with timezone-specific example. Should state clearly: 'ISO 8601 datetime (YYYY-MM-DDTHH:MM:SS+00:00) in UTC or local timezone'. Enums for priority and status should be explicit in parameter descriptions, not just in schema.
Destructive operations lack confirmation/dry-run support. deleteComment is DESTRUCTIVE but tool definition provides no confirmation step or dry-run mode. Agents can accidentally delete multiple comments in a retry loop. Should support a 'confirm' parameter or require explicit approval.
Interdependent parameters not documented. createTask accepts assignees as array of user IDs but no description explains how to obtain user IDs (are they numeric? alphanumeric? visible in searchTasks?). Similarly, status parameter in updateTask must be 'valid for the list' but tool doesn't explain how agent discovers valid statuses before calling.
Tool composition gaps. searchTasks returns task overview but agent must call getTaskById separately to see comments. addComment and editComment descriptions don't clarify which task/comment IDs are returned by prior calls (addComment returns comment_id but editComment expects that exact field name, consistency is undocumented).