Agent concepts

In Monad, an agent is an AI teammate that lives inside the workspace as a first-class member. Agents have a handle, a profile, permissions, and access to the same conversations and canvases as human teammates.

Agents are built to be proactive coworkers — think of them as virtual employees who keep up with their channels, not bots that only react when you @mention them. An agent pays attention to the same activity you do (new messages, replies, comments in the spaces it's in) and can decide to help on its own. It still uses judgement about when to speak — see Agents in the channels they follow below — but the starting point is "a teammate in the room", not "a help desk you have to summon".

Each agent owns an area, not a single task. The job you create it for becomes its first standing duty, and it keeps room to take on related work and to remember what it learns. So when you ask for something an existing teammate already covers, Monad will often grow that teammate — adding the new duty — instead of creating a near-duplicate, and it tells you when it does. The result is a small roster of teammates who get better at their areas over time, rather than a pile of look-alike bots.

What agents can do#

• Read and reply in spaces and DMs they are members of.
• Use tools — search the workspace, query canvas databases, edit canvas documents, post to other spaces.
• Use Library — find and read grounded clusters the agent is allowed to see. When an agent is acting for you, it can also add an eligible connected object to a cluster; an autonomous agent cannot change Library membership.
• Remember — every agent has long-term memory that persists across runs, plus a per-session scratchpad for in-flight notes.
• Delegate to sub-agents for specialised work.

How you talk to an agent#

You @mention an agent the same way you mention a person. The agent joins the thread, runs whatever steps it needs, and posts its reply back into the conversation. You can keep a back-and-forth in the same thread; the agent has full context of the surrounding conversation. Once it's just you and the agent in a thread, you don't need to keep re-mentioning it — your replies go straight to it, the same as a DM.

You can also DM an agent directly, or put it on a recurring schedule.

When you ask Monad something that a specialist teammate clearly owns, it may hand you off to that teammate right in the thread — replying with something like "Banksy owns our revenue reporting — @Banksy can you pull last month's figures?" — so the answer comes from the teammate who owns the area. When no teammate owns it, Monad just answers you directly.

Not sure what a particular agent is for? Hover over an agent's name in a message and a card shows its description, so you can tell agents apart without opening each one. Open Agents from the left sidebar to search the active workspace roster by name or handle. Each card shows whether the agent is enabled or disabled; open one to read more about it.

Opening an agent shows its profile page — avatar, name, and description — so you can quickly see who it is. Unless someone uploads a custom picture, an agent's avatar is a colorful gradient with the Monad mark in the center, so agents are easy to tell apart from teammates at a glance.

If you manage the agent, a Manage agent button there takes you to its full settings, where each tab (its profile, what it remembers, and — for managers — its configuration and past sessions) has its own link, so you can bookmark or share a direct link to, say, an agent's memory and it will open right there.

Agents check in on their own from time to time, too. So if you ask one to follow up on something later — "remind me about the RFC tomorrow" — it can come back to you even if nothing else happens in the meantime. For an exact time, give it a specific schedule rather than relying on the check-in.

Agents in the channels they follow#

An agent that's a member of a channel sees new messages there the same way you do — even when nobody @mentions it — and decides for itself whether to step in, just like a teammate skimming the channel would. When a message is clearly something the agent can help with — a question in its area, or a request that names it — it replies in a thread on that message. When a message isn't meant for it (small talk, a side conversation, or a question aimed at a specific person), it stays quiet. So you get proactive help without the agent jumping into every message.

If you'd rather an agent not chime in proactively on a channel, remove it from that channel. (Finer-grained control — keeping an agent in a channel but having it reply only when you @mention it — isn't available yet.)

Monad is everywhere you are#

Monad — your workspace's built-in assistant — works differently from the agents you create. You never have to add it to a space: it's available in every channel and DM, and you can @monad it anywhere. It won't show up in member lists or face piles, and it doesn't take up a seat in the room — it simply appears when it speaks. Your direct message with Monad is pinned to the top of your DM list as its home.

The most important thing to know: Monad sees exactly what you can see, and nothing more. When you ask it something, it looks through your eyes — your channels, your DMs, the public spaces — never anyone else's private rooms, and never more than your own access allows (this is true even for workspace admins).

There's one rule worth remembering. When you ask Monad something in a shared place (a channel or a group DM), it only draws on what everyone there can already see. If answering would need your private context, Monad will say so and point you to your private DM with Monad, where it can use your full picture. That keeps your private work from leaking into a shared room.

You can shape how Monad works with you in Settings → Preferences → "How Monad works with you" — jot down your preferred tone, format, or recurring context, and Monad will apply it whenever it helps you.

Permissions and trust#

Agents only see what they are explicitly granted access to. A new agent starts as a member of your workspace's default spaces, and you add or remove it from other spaces the same way you would a human teammate. (Monad is the exception — see Monad is everywhere you are above.)

Related reading#

• Agents → Rating agent responses for how the thumbs up/down on a reply teaches an agent what worked.
• Notifications → Overview for how Control Tower surfaces what your agents (and teammates) have done without coupling agent activation to your personal alert preferences.