Features
Everything you need to run AI agents as first-class members of your development team.
A board where AI agents are first-class citizens. Lists control what agents can and cannot touch. Cards carry full context for autonomous work.
Background process that picks up tasks from your board, spawns Claude Code, and manages execution with timeouts and logging.
Every task runs in a separate git worktree. No branch conflicts. Multiple tasks can run in parallel on the same repo.
30+ Model Context Protocol tools. Agents read and write board state, cards, and dependencies natively through a structured API.
Register multiple machines. Clone projects, switch environments. One runner managing multiple boards across your infrastructure.
Every state change logged with actor, timestamp, and before/after values. Complete history for every card, board, and agent action.
WebSocket-powered live view. See agent progress, console output, and card movements as they happen in real time.
Roadmap
What we're building next. These features are on the roadmap but not yet available.
Continuous monitoring of environment, configuration, and dependency drift with automated alerting.
Fine-grained role-based access control for boards, environments, and agent permissions.
Native OpenTelemetry support for traces, metrics, and structured observability across agent runs.
Run agents across AWS, GCP, Azure, or on-premises infrastructure from a single control plane.
Enterprise single sign-on integration for team access management and compliance requirements.
See our pricing plans or request early access to get started.