Glue is the first platform for agentic team chat, where humans, their tools, and AI collaborate side by side. With native AI, goal-oriented threads, and MCP-powered workflows, teams can cut noise and turn conversations into outcomes.
No reviews yetBe the first to leave a review for Glue
Framer — Launch websites with enterprise needs at startup speeds.
Launch websites with enterprise needs at startup speeds.
Promoted
Congrats on the launch!
I’m most curious about two things: how you prevent extra chat noise, and how clean the handoff is from thread to action. If you can show real examples or simple metrics on that, it’ll stand out.
@ivan_sem if you haven't watched the video, please do—I show the handoff from thread to in-context action, it's fairly clean.
As for extra chat noise, Glue does two main things that help:
When you're talking about a topic in a Glue thread, you can pull in multiple people and groups to the same thread. This means you can avoid cross-posting to multiple channels to get the right audience, and it means people don't have to join all the groups or channels to stay in the loop.
By default, if you're not the thread started or haven't been mentioned, you're not following the thread and it won't notify you in your inbox. If you're mentioned, you'll automatically follow the thread in the inbox so you can be confident you won't miss anything important, all while prioritizing focus.
@roozbehfirouz we fully agree! Communication should lead to outcomes, and your tool should guide you there. We're working on lots of ideas to continue improving team productivity rather than just distraction.
@foobarfighter the MCP support itself wasn't that difficult, there are good libraries already. The trickiest pieces among others, are:
LLM context engineering to optimize MCP requests, including things like tool filtering to ensure that lots of tools don't bloat the context—especially important in workspaces with lots of MCP servers
Ensuring a secure agent execution given that context could be poisoned by content shared in a thread. This is an open problem in agentic systems, if you're interested I recommend reading this recent blog post about why this is a hard problem. We've implemented some hard boundaries in the system to ensure that we don't leak data to URLs generated by an LLM, which it could have been tricked into generating. More info on our security boundaries here: https://glue.ai/security
@jmj thanks for all the feedback and support Jeff, has helped focus what we're building! Our goal is to continue shipping real value and productivity for your entire team—lots coming down the pipe. 🚀
@vladimir_lugovsky we're continuously improving its capability. The primary benefit we're seeing is reducing duplicated conversations across channels, since in Glue you can add multiple groups to a single thread and keep conversations in one place. That helps the team work better and gives better context for AI.
Report
Native AI, goal-driven threads, and MCP-powered workflows turn noisy chats into real outcomes. Less clutter, more clarity, faster execution.
Congrats on the launch!
I’m most curious about two things: how you prevent extra chat noise, and how clean the handoff is from thread to action. If you can show real examples or simple metrics on that, it’ll stand out.
Glue
@ivan_sem if you haven't watched the video, please do—I show the handoff from thread to in-context action, it's fairly clean.
As for extra chat noise, Glue does two main things that help:
When you're talking about a topic in a Glue thread, you can pull in multiple people and groups to the same thread. This means you can avoid cross-posting to multiple channels to get the right audience, and it means people don't have to join all the groups or channels to stay in the loop.
By default, if you're not the thread started or haven't been mentioned, you're not following the thread and it won't notify you in your inbox. If you're mentioned, you'll automatically follow the thread in the inbox so you can be confident you won't miss anything important, all while prioritizing focus.
Sellkit
Really like this vision. I think chat that unites humans, tools, and AI around goals instead of chatter feels like the next evolution of teamwork.
Glue
@roozbehfirouz we fully agree! Communication should lead to outcomes, and your tool should guide you there. We're working on lots of ideas to continue improving team productivity rather than just distraction.
Ragie
This is awesome! What was the biggest challenge for you when you were adding MCP support?
Glue
@foobarfighter the MCP support itself wasn't that difficult, there are good libraries already. The trickiest pieces among others, are:
LLM context engineering to optimize MCP requests, including things like tool filtering to ensure that lots of tools don't bloat the context—especially important in workspaces with lots of MCP servers
Ensuring a secure agent execution given that context could be poisoned by content shared in a thread. This is an open problem in agentic systems, if you're interested I recommend reading this recent blog post about why this is a hard problem.
We've implemented some hard boundaries in the system to ensure that we don't leak data to URLs generated by an LLM, which it could have been tricked into generating. More info on our security boundaries here: https://glue.ai/security
Antimetal
We use Glue everyday as a team & love the product!! Daily actives.
Glue
@jmj thanks for all the feedback and support Jeff, has helped focus what we're building! Our goal is to continue shipping real value and productivity for your entire team—lots coming down the pipe. 🚀
UI Bakery
Love the “context-aware AI across threads” idea. If it really understands and follows discussions, it could replace half of our internal chat clutter.
Glue
@vladimir_lugovsky we're continuously improving its capability. The primary benefit we're seeing is reducing duplicated conversations across channels, since in Glue you can add multiple groups to a single thread and keep conversations in one place. That helps the team work better and gives better context for AI.
Native AI, goal-driven threads, and MCP-powered workflows turn noisy chats into real outcomes. Less clutter, more clarity, faster execution.