Feature Requests

A dedicated page for where generated artifacts can be retrieved and organized. Like the Clips Panel.

Hi, We're evaluating ClickUp's AI Notetaker as a potential replacement for Microsoft CoPilot in Teams. One of the key reasons we currently use CoPilot is its ability to automatically delete meeting recordings after a configurable period (e.g. 3 months), while retaining the AI-generated notes, summary, and action items indefinitely. This gives us the best of both worlds: useful meeting intelligence without the data protection liability of storing recordings long-term. Feature request: Could a similar scheduled retention/auto-deletion policy be added to AI Notetaker? Ideally at the Workspace admin level, allowing us to set a timeframe after which recordings and transcripts are automatically purged, while the AI-generated notes (overview, key takeaways, next steps, key topics) remain intact. This would be a significant help for GDPR compliance and general data hygiene.

I have a conversation for every brand I work, but there's another chats with Brain for other random stuffs. I need a way to pin my brand chats, or use folders to group AI conversations on Brain AI.

We run our own AI agents on our own server, one per client account. We want a teammate to @mention an agent in a task comment and get the answer back in the same thread, the way the Cursor integration works today. ClickUp Support (ticket 2746783) confirmed this is not possible for custom agents: the Cursor agent is a curated App Center partnership with an agent identity built into the platform, and a bot account cannot get the same experience. The workaround is a taskCommentPosted webhook plus a filter on the bot's user id plus a reply through the API. It has one gap that breaks conversations: taskCommentPosted fires for top-level comments only, so a reply inside the thread never reaches the agent and we have to poll. The request, in order of value: A way to register a custom agent identity from the App Center or the Public API, with its own name and avatar, that members can @mention in comments and assign tasks to, the same as Cursor and Super Agents. A webhook event when that agent is mentioned, with the mention delivered as a structured block and fired for threaded replies as well as top-level comments. A way for the agent to reply in the thread under its own identity, without consuming a member seat. Why it matters: teams that already host agents elsewhere (self-hosted, OpenAI, Anthropic, or open source frameworks) want ClickUp to stay the one place where the conversation about a client happens. Super Agents and Brain do not cover this case because the model and the memory live outside ClickUp.

I would like to be able to take an artifact I have created and move it to a list or folder level.

A native 'Duplicate' option for Skills and Subskills that preserves all formatting (.md) would be useful

Currently the only way to disable AI is through ClickApps via Admin. This disables it for the entire team. I just want to remove this garbage from my own personal profile without impacting anyone else's workflow.

When chatting in Arabic or other RTL languages, Brain's responses render left-to-right, making them hard to read. Responses should detect the language direction and align text accordingly (right-to-left for Arabic, Hebrew, etc.). This would make Brain much more usable for RTL-language users.

Similar to "projects" in ChatGPT, it would be incredibly useful to allow individual conversations had with Brain placed into their own folders vs the current way of having just a long list of past conversations. This could also go a step further and allow for chats to be associated with specific tasks/docs manually or automatically based on where the user is currently working inside of ClickUp.

It would be great if Brain² had a voice feature — and while it might seem like a gimmick at first, there are genuine productivity use cases that make it worthwhile. One example is when you're working through something with AI: right now, you have to stop what you're doing to read its response. With a voice output option, you could provide your input, let Brain² process it, and then listen to the response while staying focused on something else, further helping with multitasking.