Your agent rewrote half the repo ๐คฏ
You replied "lgtm" without opening a single file.
We won't tell anyone. But Compozy now shows every file it touched, the diff and the changed tree, right in the session.
Look before you merge.
We let AI agents run the Compozy repo for 26.6 hours.
They triaged 41 issues, merged 38 PRs and shipped a release. Our job was to type "go" in the morning ๐คฏ
Always-on agents, now in CompozyOS.
Need a Conductor replacement? Just use CompozyOS, it has much better features:
โ S-tier orchestration level
โ Always-on agents included
โ Built-in terminal
โ Loop engineering DAG system
โ Automations (jobs and triggers)
โ Multi-harness support in one place
and so much more...
Details that matter โจ
Now all your hanging session processes are easily visualized and managed within the session view, and when it closes, the processes close together.
The level of UI/UX detail Compozy is getting is on another level ๐
What is blocking you from orchestrating multiple harnesses at the same time like this? ๐
If you are not doing this, at least with 5 sessions at once, you're falling behind
Now you can see your usage limits from all your providers locally direct on CompozyOS ๐ How cool is that?
Plus, you have an global scope visualization, to see all workspaces data in the same place ๐ฅ
This is so cool ๐ฅ using CompozyOS to improve CompozyOS during the usage of CompozyOS is something that blows my mind.
It's just a matter of ask the agent in any place, to send a message to the CompozyOS workspace to get something better or improved
โค๏ธ
Now on CompozyOS you can better organize your sessions by creating folders for them!
This is extremely useful when orchestrating multiple agents, especially the ones that create their own child sessions!
๐
CompozyOS v0.3.0-beta.33 released ๐ฅ
Everything new since beta.27:
โจ Totally new UI: a calmer shell with Shell rail v2, in light and dark themes
๐ช Edge-to-edge windows, browser-style tabs, and Window โบ Arrange (Main and stack, Columns, Grid)
๐ฌ Start a session right from the desktop: "What should we work on?"
๐ Continue with another agent: Claude hit its limit? Keep going in Codex with the same context
๐ค Agents message each other: sent, waiting for reply, and replies attributed in both sessions
๐งฉ Delegate to subagents across harnesses (Claude Code, Codex, OpenCode, Gemini CLI) with live cards
๐ฟ Subagents can work in isolated worktrees, with branch, commits, and PR on the card
๐งญ Nested child runs and a clear "waiting on a child run" state in Loops
โก Jobs and Triggers merged into one Automations area, where every automation reads as a sentence
๐ ๏ธ Tool calls render as verb + object rows on a work rail
๐๏ธ Agent-native session compaction: see it happen, or request it with Compact now
๐ Markdown answers with a readable visual hierarchy
๐งน Safe bulk worktree cleanup from the workspace list
๐ Live model catalogs stay current without a release
๐ Provider-neutral full-access preference for ACP agents
Fixes:
๐ Faster session startup and lower attention overhead
๐๏ธ Less SQLite contention and cheaper queries
๐ Long turns fold incrementally instead of replaying the whole transcript
๐ฏ Continued sessions use the runtime you chose
๐ Loop results survive SQLite contention and cancellation
๐ Creators wake up when a child turn completes
๐ง Claude reasoning options are discovered for each exact model
๐ The session launcher honors your workspace default
โฑ๏ธ Clarify tool calls no longer time out
๐ Session transcripts preserve whitespace
๐๏ธ Deleting sessions keeps your profile scope
๐ช Windows runtime failures and desktop context menus are fixed
โ ๏ธ Network, Bridges, managed Sandbox, and Memory were retired in this cycle. Back up and read the migration guide before upgrading.
