Features
Visualize entire agent context as turn-based sections
Fold sections into summaries to free context
Unfold sections back to full detail instantly
Pin sections to prevent automatic folding
Peek inside folded sections without changing agent context
Automated Conductor folds cold, unfolds relevant context
Protected working tail never folded (default ~20k tokens)
Token count display for budget monitoring
Every action attributed and replayable
Cross-platform: macOS, Windows, Linux
Open source under MIT license
Built for integration with Claude Code and coding agents
Laguna XS.2 model (33B params, 3B active, on-device)
Laguna M.1 model (225B params, 23B active, via API)
256K context length for multi-step reasoning
Multi-agent orchestration with planning and tool use
Sandboxed agent execution environments
Developer surfaces: agents, TUI, IDE extensions, binaries
Custom foundation models deployed on-prem or in VPC
Data connectors to repos, databases, data warehouses
Role-based access control for humans and agents
Executive-grade governance and auditability
On-prem, VPC, and workstation deployment (defense only)
Forward Deployed Research Engineers embed with teams
Real-time observability with end-to-end traces
Air-gapped network support
Open-weight model weights available
