Teamily AI - Your personal agent, agent teams & people in one messenger
by•
The human-agent platform where people and always-on agents get work done together in one unified, context-aware space. Unlike single-user personal agents, Teamily AI enables multi-user, multi-agent teamwork. Built on an AI-native messenger, it connects people, agents, docs, meetings, calls, workflows, and business systems to automate office work and customer service. Agents become real teammates with persistent memory, proactive intelligence, and continuous self-improvement. Free to get started.
Replies
Can it plan an event from venue docs, links, and photos?
@mi_leeli Yes, absolutely! Teamily supports both multimodal and context-aware prompting. You can share multiple documents, links, and photos across different messages with an agent or in a group chat. Then, simply ask the agent to plan an event for you. It will use the entire conversation history and shared memory to understand the context and create a personalized plan.
No need to put everything into a single prompt. Just share information naturally as you chat, and let Teamily handle the rest. Give it a try!
@mi_leeli Yes, Drop the venue docs, links, and photos into the group, @ mention an agent, and it pulls everything into one plan: options, timeline, and a summary page you can share with everyone. The whole group can then review and tweak it together in the thread.
@mi_leeli Mi event planning is such a good one!! Venue docs + photos + everyone's opinions is basically the perfect agent job 😄 And because the plan lives in the group, when someone says "wait, what about parking?" it can update right there instead of starting a new doc. Would love to see what you plan with it!
Can I upload a long video and ask questions about one section?
@vilmos_marton yes. Please have a try.
@vilmos_marton Yes, Vilmos! Drop a long video into the chat and ask about a specific part, for example, "What did they say about pricing around the 20-minute mark?" The agent answers in the thread, so your whole team sees it too.
@vilmos_marton Mi event planning is such a good one!! Venue docs + photos + everyone's opinions is basically the perfect agent job 😄 And because the plan lives in the group, when someone says "wait, what about parking?" it can update right there instead of starting a new doc. Would love to see what you plan with it!
Congrats!
Roxie's question about how proactive an agent should be in a group of ten got me thinking. What happens when two people give it conflicting instructions? Does it decide, or ask the group?
@genedai Great question💡, Gene, and one we think about a lot. Our principle is that humans steer, and agents shouldn't silently pick a side. When instructions conflict, the agent surfaces it to the group, lays out both options, and asks who should decide, instead of guessing. Agents can push back on each other too, but the final call stays with the people in the room. Congrats on Metix AI as well!
@genedai Ha, you picked the one I lose sleep over. Like Xavier said, the agent shouldn't quietly pick a side. The part I'm still chewing on is how it says so without making two teammates feel called out in front of everyone.
the first place I would try this is a messy planning group😼
@jody_l_wyatt A messy planning group is the perfect place to start 😄 Drop in everyone's links, photos, and ideas, @ mention an agent, and it pulls it all together into one plan the whole group can see and tweak. No more scrolling back 200 messages to find who said what.
Does the agent keep working if everyone closes the app?
@subhash_kanagamani Yes, Subhash! That's the whole idea behind always-on agents. They run in the cloud, so they keep working after everyone logs off: finishing research, following up on email, or running recurring tasks on a schedule. When you come back, the results are waiting in the thread. Your agent team doesn't sleep 🌙
@subhash_kanagamani yup!
I like the idea of seeing the work happen where the request started 👍
@orman_canida Thanks Orman! That's exactly it. The request, the work, and the result all live in the same thread, so everyone sees how it happened and can jump in at any point.
@orman_canida Yes, if while you're trying it out there are any features you'd like us to add to the product, you can send those over to us as well. We always love seeing real user needs 😄
It is hard to review AI work when everyone is looking at a different version.
@mengqi_jin So true. A "looks good" means nothing if nobody knows which version it was about. Keeping feedback pinned to the exact draft is one of the things I care most about getting right.
so，can I see what context was used before an email gets drafted
@tomzoe You should be able to, and that's very much on my list. For now you can ask the agent what it's going off before it drafts, and have it wait for your OK before anything gets sent.
@tomzoe you can see the "thinking" of the agent to observe the entire process it used for doing the task ...
Could the revised deck stay attached to the design review thread?
A private client group with one writing agent could be a good first test.