Comment threads can now be made private to certain users, allowing you to create internal-only notes, private review workflows, and feedback that only the right people can see. Plus, adding AI to comment threads is easier than ever.
Join thousands of companies using Liveblocks to build multiplayer experiences for people and agents.
Most companies are trying to figure out how people and AI can naturally work together. Our opinion on how humans and AI will work together is that the best place to start is with patterns people already understand, such as comments. They’re familiar, they’re anchored to real context, and they turn out to be one of the most natural ways to bring AI into a product.
Check out what we’ve recently released, whether you already have Comments live or have been meaning to add them.
Comment threads now have a visibility property. Each thread is public by
default, private when you need it, with permissions to control exactly who can
see and reply. That unlocks internal-only notes, private review workflows, and
feedback that only the right people can see.
Creating a private comment in our Private Commenting example.
To enable creating private threads, pass the visibility="private" option to
Composer.
To hide private threads from certain users, ensure they cannot see them with the
"comments:private:none" permission. For example, when creating a room, you can
create a group of users that has no access to see them. Below, admin users
will see all threads, whereas regular users will not be able to see private
threads
To get started with this feature, follow our full guide on how to adding private commenting to your app. We also have a live example that shows how to use it.
A comment thread is anchored to a specific place in your app. So when someone leaves a comment that says “make this bigger,” there’s no ambiguity about what “this” is—and any AI working in the thread has the full conversation as context.
Add an AI assistant as a user people can @ mention right in the composer, and
a comment becomes a way to put agents to work alongside your team.
A contextual comment in our Collaborative Flowchart AI example
To enabling AI replies, you need to add an AI user to your app, the same way you would add a human user.
You can then listen for
commentCreated webhook events,
check if the AI user was mentioned, and generate a response if they were.
There are two ways to add AI replies to your existing Comments app:
We’ve expanded our get started guides with a range of new examples, so you can drop in ready-made, contextual commenting and have it working in minutes, not weeks.
Get started with Comments
