clickup-cli
ClickUp CLI for managing tasks, sprints, comments, and statuses. Use when the user needs to interact with ClickUp — creating/editing tasks, checking sprint status, adding comments, linking PRs, or searching tasks. Prefer this CLI over raw API calls.
SKILL.md
| Name | clickup-cli |
| Description | ClickUp CLI for managing tasks, sprints, comments, and statuses. Use when the user needs to interact with ClickUp — creating/editing tasks, checking sprint status, adding comments, linking PRs, or searching tasks. Prefer this CLI over raw API calls. |
name: clickup-cli description: ClickUp CLI for managing tasks, sprints, comments, statuses, and Docs. Use when the user needs to interact with ClickUp — creating/editing tasks, checking sprint status, adding comments, linking PRs, managing Docs and pages, or searching tasks. Prefer this CLI over raw API calls.
ClickUp CLI (clickup)
Use the clickup CLI instead of raw ClickUp API calls. It handles authentication, git integration, fuzzy status matching, and custom fields automatically.
When to Use
- User asks to create, edit, view, or search ClickUp tasks
- User wants to check sprint status or recent tasks
- User needs to add comments, link PRs/branches, or manage task statuses
- User mentions ClickUp task IDs (e.g.,
CU-abc123,86abc123) - User asks about their ClickUp inbox or mentions
Authentication
clickup auth login # Authenticate with API token
clickup auth status # Check auth status
Configuration is stored in ~/.config/clickup/config.yml. Supports per-directory defaults for space, team, folder, and list.
Task Management
View & Search
# View a task (auto-detects from git branch if no ID given)
# Also detects tasks via PR URL in descriptions when branch has no task ID
# Subtask lines show due/start dates inline for at-a-glance visibility
clickup task view
clickup task view CU-abc123
# View with JSON — useful for extracting subtask IDs for bulk operations
clickup task view 86abc123 --json
# Bulk view: fetch multiple tasks concurrently (up to 10 parallel)
clickup task view 86abc1 86abc2 86abc3 --json
# Extract tags from multiple tasks
clickup task view 86abc1 86abc2 86abc3 --jq '[.[] | {id: .id, tags: [.tags[].name]}]'
# Search tasks by name and description (supports fuzzy matching)
# Uses server-side search (Level 0) with parallel space traversal
clickup task search "login bug"
clickup task search "login bug" --exact # Exact matches only
clickup task search "login bug" --assignee me # Only your tasks
clickup task search "login bug" --assignee "alice" # By name/username/ID
clickup task search "Phase 1" --include-subtasks # Discover subtasks by name without knowing parent ID
# Recent tasks (excludes archived folders)
clickup task recent
clickup task recent --sprint # Only current sprint tasks
# List tasks in a specific list
clickup task list --list-id 12345
# List tasks using configured default list (via `list select`)
clickup task list
# Include closed tasks in the list
clickup task list --include-closed
clickup task list -c --list-id 12345
# Include subtasks in the list (off by default)
clickup task list --include-subtasks --list-id 12345
# Recursive view — fetch subtasks and their children in a single tree
clickup task view 86abc123 --recursive --json
# Bulk delete tasks (requires confirmation or -y to skip)
clickup task delete 86abc1 86abc2 86abc3 -y
# Task activity/comment history
clickup task activity CU-abc123
Navigating task hierarchies: Use --json to drill into subtask trees. The JSON output includes a subtasks array with each subtask's id, name, status, due_date, and start_date. To operate on subtasks in bulk, view the parent with --json, extract the subtask IDs, then pass them to task edit.
Task Naming Conventions
IMPORTANT: Always use descriptive, well-structured task names. Task names should be scannable on a sprint board without needing to click into the task. Follow this convention:
Format: [Work Type] Context — Action (Platform/Scope)
| Component | Purpose | Examples |
|---|---|---|
[Work Type] | Categorises the work at a glance | [Packdown], [Bug], [Feature], [Refactor], [Spike], [Hotfix] |
Context | Campaign, project, or feature name | NT x THL, Auth v2, Booking Flow |
Action | What is being done — use imperative verbs | Remove landing page, Add SSO support, Fix timeout error |
(Platform/Scope) | Which app, site, or system is affected | (CamperMate), (Britz + Apollo), (API), (CamperMate + CM.com) |
Examples:
[Packdown] NT x THL — Remove CamperMate campaign landing page[Packdown] NT x THL — Remove app menu links (CamperMate + CM.com)[Bug] Booking Flow — Fix timeout on slow connections (API)[Feature] Auth v2 — Add SSO support (CamperMate)[Spike] Search — Evaluate Algolia vs Typesense
When to simplify: For small, self-evident tasks (e.g. subtasks or checklists), a short descriptive name is fine. The full convention is most valuable for sprint-level tasks that need to be scannable across teams.
Template tasks: Use {Campaign} or {Project} as a placeholder in the context position so the name can be filled in when the template is instantiated (e.g. [Packdown] {Campaign} — Remove campaign landing page).
Create & Edit
IMPORTANT: When creating a task, fill in ALL applicable fields. Do not create bare tasks with just a name. Ask the user for any information you don't already have. Use the naming conventions above for the --name flag. A well-created task should include as many of these as possible:
--current— preferred: auto-resolves the active sprint list (no need to know the list ID)--list-id— explicit list ID (use when creating outside the current sprint)--list-name— resolve a list name to its ID (e.g.,--list-name "Sprint 89")--name— task name (required — follow naming conventions above)--descriptionor--markdown-description— clear description of the work--status— initial status (e.g., "open", "in progress")--priority— 1=Urgent, 2=High, 3=Normal, 4=Low--assignee— user ID(s) for who should work on it--tags— relevant tags (repeatable)--due-date— deadline (YYYY-MM-DD)--start-date— when work should begin (YYYY-MM-DD)--time-estimate— estimated effort (e.g., "2h", "4h", "1d")--points— sprint/story points--parent— parent task ID if this is a subtask--links-to— related task ID--type— 0=task, 1=milestone--field "Name=value"— custom fields (repeatable)
IMPORTANT: Always use --current instead of hardcoding sprint list IDs. Sprint list IDs change every sprint. Never cache or remember a specific list ID — always let the CLI resolve it dynamically.
After creating a task, consider adding checklists for acceptance criteria or subtasks:
clickup task checklist add <task-id> "Acceptance Criteria"
clickup task checklist item add <checklist-id> "Unit tests pass"
clickup task checklist item add <checklist-id> "Code reviewed"
clickup task checklist item add <checklist-id> "Assigned item" --assignee 12345678
Example — create a task by list name (resolves name to ID):
clickup task create --list-name "Sprint 89" --name "Fix bug" --status "open"
Example of a well-populated task creation:
clickup task create --current \
--name "[Bug] Auth — Fix login timeout on slow connections (API)" \
--markdown-description "Users on slow 3G connections get a timeout error..." \
--status "open" \
--priority 2 \
--assignee 12345678 \
--tags "bug" --tags "auth" \
--due-date 2025-03-01 \
--start-date 2025-02-20 \
--time-estimate 4h \
--points 3
Tags: Before creating a task, check what tags are available in the space with clickup tag list. Use existing tags rather than inventing new ones — tags should be consistent across the workspace. If no suitable tag exists, discuss with the user before creating a new one.
Bulk create from JSON file — use --from-file for batch task creation:
# Create many tasks at once from a JSON file
clickup task create --current --from-file tasks.json
clickup task create --current --from-file tasks.json --json
The JSON file should contain an array of task objects:
[
{
"name": "Design homepage",
"description": "Create wireframes and mockups",
"status": "open",
"priority": 2,
"due_date": "2026-03-15",
"time_estimate": "4h",
"tags": ["design"],
"parent": "86abc123",
"fields": [{"name": "Environment", "value": "staging"}]
},
{
"name": "Implement auth flow",
"start_date": "2026-03-01",
"due_date": "2026-03-10",
"assignees": [12345678],
"points": 5
}
]
Errors on individual tasks are reported but don't stop the batch. A summary is printed at the end.
# Edit a task (auto-detects from git branch)
clickup task edit --status "in progress" --priority 2
clickup task edit CU-abc123 --field "Environment=production"
clickup task edit --due-date 2025-03-01 --time-estimate 4h
# Bulk edit multiple tasks at once — pass all IDs as positional args
clickup task edit 86abc1 86abc2 86abc3 --status "Closed"
clickup task edit 86abc1 86abc2 86abc3 --due-date 2026-03-01 --priority 2
# Tags — add without removing existing
clickup task edit CU-abc123 --add-tags new-feature-development
clickup task edit 86abc1 86abc2 --add-tags r&d,new-app-development
# Tags — remove specific tags
clickup task edit CU-abc123 --remove-tags fix
# Tags — replace all (use with caution)
clickup task edit CU-abc123 --tags "ios,android,new-app-development"
# Custom fields
clickup task edit CU-abc123 --field "Environment=production"
clickup task edit CU-abc123 --clear-field "Environment"
clickup field list --list-id 12345 # Discover available fields
Bulk edit workflow — closing all subtasks of a parent:
- View the parent task with
--jsonto get subtask IDs:clickup task view 86parent --json # → .subtasks[].id gives you the child IDs - If subtasks themselves have children, view each subtask with
--jsonto get the next level of IDs. - Pass all collected IDs to a single
task edit:clickup task edit 86child1 86child2 86child3 ... --status "Closed" - Errors on individual tasks are reported but don't stop the batch. A summary line shows
Updated X/N tasks.
Important: Check clickup status list or the validation error output to find the correct status name for the space — statuses vary between spaces (e.g., "done" vs "Closed" vs "complete").
Multi-List (Add/Remove Tasks from Lists)
# Add a task to a sprint list (task stays in its original list too)
clickup task list-add 86abc123 --list-id 901613544162
# Add multiple tasks to the same list at once
clickup task list-add 86abc1 86abc2 86abc3 --list-id 901613544162
# Remove a task from a list (must belong to at least one other list)
clickup task list-remove 86abc123 --list-id 901613544162
# Remove multiple tasks from the same list
clickup task list-remove 86abc1 86abc2 86abc3 --list-id 901613544162
Use list-add when a task needs to appear in multiple lists — e.g., a campaign task that also belongs in an engineering sprint. Use list-remove to undo this. Neither command moves the task; they manage secondary list memberships.
Status Management
# Set status (supports fuzzy matching: "review" matches "code review")
clickup status set "in progress"
clickup status set "in progress" CU-abc123
# List available statuses
clickup status list
clickup status list --space 12345
# Add a new status to a space (inserted before Closed)
clickup status add "done"
clickup status add "QA Review" --color "#7C4DFF"
clickup status add "done" -y # Skip confirmation
clickup status add "done" --space 12345 # Specific space
Status values are fuzzy-matched: exact match > contains match > fuzzy match. If ambiguous, the CLI picks the most specific match and prints a warning.
Guardrails for status add: Only suggest adding a status when there's a genuine gap in the workflow (e.g., a space has "Closed" but no "done" equivalent). Always confirm with the user before adding. Never remove statuses without explicit user instruction — statuses affect all tasks in the space.
Sprints
# Show current sprint tasks
clickup sprint current
# List all sprints in a folder
clickup sprint list
Folders
# List folders in a space
clickup folder list
clickup folder list --space 12345
clickup folder list --archived --json
# Select a default folder (interactive)
clickup folder select
clickup folder select --local # per-directory
Lists
# List lists in a folder
clickup list list --folder 12345
clickup list list --space 12345 # folderless lists
clickup list list --json
# Select a default list (interactive)
clickup list select
clickup list select --local # per-directory
Comments
comment add, comment edit, and comment reply all parse the body as
markdown — headers (##), bold (**x**), italic (*x*), inline code,
fenced code blocks, ordered/bullet lists (with nesting), blockquotes, and
links all render as rich formatting in ClickUp.
@username resolves against workspace members (case-insensitive) and also
accepts the first-name token or email local-part when unambiguous —
so @alice works for "Alice Smith" if she's the only Alice in the workspace.
# Add a comment (markdown + @mention shortcuts)
clickup comment add CU-abc123 "Looks good, @alice please review"
clickup comment add CU-abc123 "## Recap
- Fixed the timeout bug
- Added regression test"
# Reply in a thread (same markdown + mention rules)
clickup comment reply <comment-id> "@bob confirmed this on staging"
# List comments (newest first — .[0] is most recent, .[-1] is oldest)
clickup comment list CU-abc123
# Edit/delete comments (you can only edit comments you authored)
clickup comment edit <comment-id> "Updated text with **bold** and `code`"
clickup comment delete <comment-id>
Git & GitHub Integration
The CLI auto-detects task IDs from git branch names. Branch naming convention: feature/CU-abc123-description or CU-abc123/description.
# Link a GitHub PR to a ClickUp task
clickup link pr
clickup link pr --task CU-abc123
# Link a specific PR number to a task (useful after merging)
clickup link pr 42 --task CU-abc123
# Link current branch
clickup link branch
# Link a commit
clickup link commit
# Sync ClickUp task info to GitHub PR description
clickup link sync
clickup link sync --task CU-abc123
clickup link sync 42 --repo owner/repo --task CU-abc123
Links are stored in the task's markdown description as rich-text with clickable URLs.
Note: When --task is specified but no PR number, the CLI first tries the current branch's PR, then searches for PRs matching the task ID in their branch name. This works even after merging when the feature branch is deleted.
Auto-detection: task view can detect the associated ClickUp task even on branches without task IDs by finding the branch's GitHub PR URL in task descriptions.
Time Tracking
# Log time to a task (auto-detects from git branch)
clickup task time log --duration 2h
clickup task time log 86abc123 --duration 1h30m --description "Implemented auth flow"
# Log time for a specific date
clickup task time log --duration 45m --date 2025-01-15
# Log billable time
clickup task time log --duration 3h --billable
# Log time for another team member
clickup task time log 86abc123 --duration 2h --assignee 54874661
# Bulk log from a JSON file
clickup task time log --from-file entries.json
Bulk time log file format:
[
{"task_id": "86abc123", "duration": "2h", "date": "2026-03-15", "description": "Feature work", "assignee": "54874661"},
{"task_id": "86abc456", "duration": "1h30m", "date": "2026-03-15", "description": "Code review"}
]
Each entry supports: task_id (required), duration (required), date, description, assignee, billable. The --assignee flag applies as a default for entries without their own assignee.
# List time entries for a task
clickup task time list
clickup task time list 86abc123
clickup task time list 86abc123 --json
# Timesheet: list all your time entries for a date range
clickup task time list --start-date 2026-02-01 --end-date 2026-02-28
clickup task time list --start-date 2026-02-01 --end-date 2026-02-28 --json
# Timesheet for a specific user
clickup task time list --start-date 2026-02-01 --end-date 2026-02-28 --assignee 54695018
# Timesheet for multiple users (fetched concurrently)
clickup task time list --start-date 2026-03-01 --end-date 2026-03-31 --assignee 48884897,54874661,54874662
# Timesheet for all workspace members
clickup task time list --start-date 2026-02-01 --end-date 2026-02-28 --assignee all
# Include task tags in JSON output (fetches concurrently)
clickup task time list --start-date 2026-03-01 --end-date 2026-03-31 --include-tags --json
When --start-date and --end-date are provided, the command switches to timesheet mode — querying all time entries across tasks for the date range, grouped by task. Defaults to the current user; use --assignee all for everyone, --assignee <user-id> for a specific person, or --assignee id1,id2,id3 for multiple users (fetched concurrently).
Use --include-tags with --json to embed task tags in the output — useful for CapEx auditing without a separate bulk-view step.
Inbox
# Show @mentions from the last 7 days (scans comments and task descriptions)
clickup inbox
clickup inbox --days 30
clickup inbox --limit 500 # Scan more tasks in busy workspaces
Workspace
# List workspace members
clickup member list
# List/select spaces
clickup space list
clickup space select # Set default space
Docs
Manage ClickUp Docs and their pages via the v3 API. All read commands support --json, --jq, and --template.
API limitation: Delete, archive, and restore operations for Docs and Pages are not available via the public ClickUp v3 API at this time.
List & View
# List all Docs in the workspace
clickup doc list
clickup doc list --json
# Filter by parent location
clickup doc list --parent-id 123456 --parent-type SPACE
# Include archived or deleted Docs
clickup doc list --archived
clickup doc list --deleted
# Paginate results
clickup doc list --limit 20 --cursor <cursor>
# View a specific Doc
clickup doc view <doc-id>
clickup doc view <doc-id> --json
Create
# Create a Doc (creates an initial empty page by default)
clickup doc create --name "Project Runbook"
# Create in a specific space with public visibility
clickup doc create --name "Team Wiki" \
--parent-id 123456 --parent-type SPACE \
--visibility PUBLIC
# Create without an initial page
clickup doc create --name "Drafts" --create-page=false
Pages
# List pages in a Doc (returns a tree structure)
clickup doc page list <doc-id>
clickup doc page list <doc-id> --max-depth 0 # Top-level only
clickup doc page list <doc-id> --json
# View a page (with optional content format)
clickup doc page view <doc-id> <page-id>
clickup doc page view <doc-id> <page-id> --content-format text/md
clickup doc page view <doc-id> <page-id> --json
# Create a page
clickup doc page create <doc-id> --name "Introduction"
clickup doc page create <doc-id> --name "Setup Guide" \
--content "# Setup\n\nFollow these steps..." \
--content-format text/md
# Create a nested page
clickup doc page create <doc-id> --name "Advanced Config" \
--parent-page-id <parent-page-id>
# Edit a page (replace, append, or prepend content)
clickup doc page edit <doc-id> <page-id> --content "New content"
clickup doc page edit <doc-id> <page-id> \
--content "## Release Notes\n\n- Fixed bug X" \
--content-edit-mode append
clickup doc page edit <doc-id> <page-id> --name "Updated Title"
JSON-first agent usage
# Get all doc IDs
clickup doc list --jq '.[].id'
# Get page IDs for a specific doc
clickup doc page list <doc-id> --jq '.pages[].id'
# Append release notes programmatically
clickup doc page edit <doc-id> <page-id> \
--content "## v1.2.3\n\n- Fixed auth timeout" \
--content-edit-mode append \
--content-format text/md
# Create doc and capture its ID
clickup doc create --name "Sprint Retro" --json | jq -r '.id'
Chat
# Send a message to a chat channel
clickup chat send <channel-id> "Hello team!"
clickup chat send <channel-id> "Deploy complete" --json
Checklists
# Add a checklist to a task
clickup task checklist add <task-id> "Acceptance Criteria"
# Add items to a checklist
clickup task checklist item add <checklist-id> "Unit tests pass"
clickup task checklist item add <checklist-id> "Assigned item" --assignee 12345678
# Edit checklist items (supports bulk — multiple item IDs)
clickup task checklist item edit <checklist-id> <item-id> --assignee 12345678
clickup task checklist item edit <checklist-id> <item-id1> <item-id2> --assignee 12345678
Common Flags
| Flag | Description |
|---|---|
--json | Output as JSON |
--jq <expr> | Filter JSON with jq expression |
--raw, -r | Output raw strings instead of JSON-encoded (use with --jq) |
--template <tmpl> | Format with Go template |
Key Behaviors
- Auto-detection: Most commands auto-detect task ID from the current git branch
- Fuzzy status matching: Status values are fuzzy-matched against available statuses
- Status validation:
task createandtask editvalidate statuses against the space's configured statuses - Archive filtering:
task recentautomatically excludes tasks from archived folders - Custom IDs: Supports both native IDs and custom IDs (e.g.,
CU-abc123) - @mentions:
comment add/edit/replyresolve@usernameto real ClickUp user tags. Accepts full username, first-name token, or email local-part when unambiguous in the workspace - Markdown comments: comment bodies are parsed as markdown and posted as rich-formatted blocks (headers, bold, lists, code, links) — plain text still posts as plain text
- Subtask discovery:
task searchandtask listaccept--include-subtasksso you can find subtasks by name without already knowing the parent task ID - Bulk operations:
task create --from-filecreates many tasks from JSON;task edit ID1 ID2 ...applies the same changes to multiple tasks. To bulk-edit subtasks: view the parent with--json, extract subtask IDs from.subtasks[].id, then pass them all totask edit - Subtask visibility:
task viewshows subtask due/start dates inline, so you can spot-check deadlines without viewing each subtask individually - Multi-list:
task list-add/task list-removemanage secondary list memberships — useful for cross-team sprint planning - Naming conventions: Task names follow
[Work Type] Context — Action (Platform)format for sprint-board scannability. Check existing tasks in the list for the prevailing convention before creating - Tag reuse: Always check available tags with
clickup tag listbefore creating tasks. Use existing tags for consistency; don't invent new ones without user confirmation - Per-directory config:
folder select --localandlist select --localstore defaults in the current directory, useful for monorepos with different ClickUp contexts - Server-side search:
task searchuses ClickUp's server-side search with parallel space traversal for faster results - Assignee shortcut:
task search --assignee mefilters results to the authenticated user; also accepts names, usernames, or IDs - Contextual task list:
task listfalls back to the configured default list (vialist select) when no--list-idis given - Bulk delete:
task delete ID1 ID2 ID3 -ydeletes multiple tasks in one command