ConvaaStart free
← All posts

Runner: code and agent actions

Let Convaa run commands and edit files on a machine you control, with the whole team watching.

Runner: code and agent actions is now available in Convaa.

Runner connects a thread to a machine you control so the model can run commands, read repositories and edit files, rather than only describing what it would do.

Because the thread is shared, agent work is visible as it happens. Everyone sees which commands ran and what came back, which is a materially different safety posture from an agent working alone in someone's terminal.

It is on the Pro plan. You can try Convaa with no account — open a thread and start right away.

Read the feature: Runner: code and agent actions →

More from the blog

Talk it through, and keep the transcript where the work is
Start a call from the thread you're already in, and the transcript is waiting there when it ends. Cameras work, screen share never, and it can run on your own hardware.
Meeting transcription that runs on your own hardware
The media server, the transcription and the audio all stay on your host. The whisper sidecar makes no outbound call, and the model is delivered with the deployment.
Running Convaa with no outbound network at all
The data plane never calls out. Point model inference at your own endpoint, ship the images in as a tarball, and the whole stack runs with no egress.
Start free, no card