AI in regulated & security-conscious workflows

AI is only usable in regulated environments if humans stay in control. Rysh is built that way: every sensitive action is approved by a person, shared output is redacted, and the whole backend can run on your own infrastructure.

The problem

Autonomous agents that can run commands are powerful — and risky. In regulated or security-conscious settings you need provable human oversight, no leaked secrets, isolation between teams, and the option to keep everything on-prem.

How Rysh helps

Every tool call that touches the system pauses for a human decision — approve once, approve always for that tool, or reject with a reason the agent must respect. Loop detection and last-prompt-wins prevent runaway behaviour. When a pane is shared, output passes through a secret-redaction layer before anyone else can see it.

The cloud backend (auth, workspaces, sharing, billing) is a Go + PostgreSQL + NATS + nginx stack that deploys via Docker Compose — so you can self-host it inside your own perimeter, with per-workspace isolation.

Key capabilities

• Human-in-the-loop approval on every sensitive tool call
• Loop detection & last-prompt-wins to stop runaway agents
• Secret redaction on shared pane output
• Per-workspace isolation on the NATS backbone
• Self-hostable server (Docker Compose: Go + PostgreSQL + NATS + nginx)
• Bring-your-own model API key; approvals keep the human in charge

See it in action

Get started

Every pane is a shell and an AI agent. Try it — install takes one command.