ClickUp Brain2: Workspace Context, Agents, Pricing, and Governance

Originally published: 2026-06-27

Last verified: 2026-07-24

TL;DR: ClickUp introduced Brain² as a company-aware AI layer over tasks, Docs, Chat, Calendar, connected apps, models, and agents. The current product pages use the names ClickUp Brain AI and Brain MAX while retaining the same core buying question: can the system answer and act with the right workspace context without crossing permissions, consuming an unclear credit budget, or creating work that nobody reviews?

What Brain uses

ClickUp describes Brain as a context-aware coworker available inside the workspace. Depending on plan and role, it can answer questions about tasks and Docs, summarize work, draft content, search connected information, and support reusable skills or agents. Brain MAX extends the experience to desktop, mobile, and browser surfaces.

The advantage over a general chat tool is access to work context. That also creates the central risk: the answer can only be trusted if the underlying permissions, sources, and action boundaries are correct.

Roles and permissions

ClickUp’s current help page says users can access Brain in locations and items they are already allowed to view. Some features are restricted by role. Members, administrators, and owners can use more agent and skill functions than guests or limited members, and editing agent permissions requires a paid plan.

A workspace pilot should test several real identities. Ask the same question as an owner, a member, a guest, and a person without access to the source project. Check the answer, cited location, and any generated output. Do not rely on an administrator account as the only test.

Agents require separate controls

Super Agents and Autopilot Agents can perform multi-step work or respond to events. They should have a named owner, a narrow trigger, a clear output destination, and a stop condition. Start with a reversible task, such as preparing a draft status report, rather than changing due dates or messaging customers.

Review visibility is also important. ClickUp notes that agent responses are visible to people who can access the response or content. Teams should decide where agent output belongs, who can approve it, and what happens when the agent cannot find enough evidence.

Follow related AI News for later ClickUp changes, but use the live help center for current role and feature rules.

Current pricing structure

The live pricing page shows a free trial layer plus Brain AI and Everything AI subscriptions. It currently displays Brain AI at $9 per user per month alongside a higher crossed-out price, and Everything AI at $28 alongside a higher crossed-out price. Buyers should confirm billing cadence, promotion terms, and whether every paid workspace member must be covered.

Advanced features may also consume AI Super Credits. The current page lists packs of 10,000 credits for $10 and separate add-ons for Talk to Text and AI Notetaker. Feature availability and limits can change, and ClickUp’s help page says future restrictions may apply.

Operational ownership

A Brain rollout needs more than a workspace setting. Name the administrator who approves models and agents, the project owners who review output, and the finance owner who monitors subscriptions and credits. Keep an inventory of active agents, their triggers, data sources, destinations, and last review date. Remove unused agents and skills rather than allowing old automations to continue after a team or process changes.

Evaluation checklist

Choose one project with clean permissions and a known source set.
Run the same questions across owner, member, guest, and denied-access roles.
Check citations, missing context, generated actions, and output visibility.
Give one agent a reversible task with a named approver and stop condition.
Record subscription cost, AI Super Credits, correction time, and failures.
Disable the pilot if a user sees data or actions outside the approved scope.