For a while we've seen companies focused solely on the coding agent part, and specially building harnesses. We thought it would be interesting to go in another direction: building on top of them.
Building on top of them allowed us to create a system that can be plugged where you work (e.g. Slack, Linear, Github, Gitlab, email, etc) and work as a peer. It really feels like a new teammate that works with you.
Some of the core components of it are:
1. Plug in any harness: codex, Claude Code, Claude managed agents, Devin, pi, etc.;
2. Memory layer. It's an abstraction, hence you may build your own implementing an interface;
3. Communication layer where you work;
4. Multi agent architecture with single lead (we started with any agent as the entrypoint and we saw lead only was better).
We are building it in public, and today we just hit 400 stars (wow). Also we have some companies that are trying to integrate it into their teams (+50 people) with promising results.
I'm really curious what you think about it, and how do you see this pattern emerging and being used by companies that want to go truely agent-native.
Any feedback or suggestions are more than appreciated!
tarasyarema•1h ago
For a while we've seen companies focused solely on the coding agent part, and specially building harnesses. We thought it would be interesting to go in another direction: building on top of them.
Building on top of them allowed us to create a system that can be plugged where you work (e.g. Slack, Linear, Github, Gitlab, email, etc) and work as a peer. It really feels like a new teammate that works with you.
Some of the core components of it are:
1. Plug in any harness: codex, Claude Code, Claude managed agents, Devin, pi, etc.; 2. Memory layer. It's an abstraction, hence you may build your own implementing an interface; 3. Communication layer where you work; 4. Multi agent architecture with single lead (we started with any agent as the entrypoint and we saw lead only was better).
We are building it in public, and today we just hit 400 stars (wow). Also we have some companies that are trying to integrate it into their teams (+50 people) with promising results.
I'm really curious what you think about it, and how do you see this pattern emerging and being used by companies that want to go truely agent-native.
Any feedback or suggestions are more than appreciated!
Cheers,