Features
Self-hosted on your own devices
Single binary, no Node.js or Python dependencies
Built-in Kanban task board for queuing work
Smart inbox to collect agent results
30+ pre-built assistants (coding, writing, research, automation)
Extensible via Hooks, MCP, Processes, and Skills
Built-in GraphRAG hybrid search
Secure sandbox for code execution
JSON/YAML DSL for agent logic
Agent Smith: describe an AI expert to build it automatically
Settings Panel to connect custom AI models (e.g., DeepSeek V4)
Native MCP support (Process, STDIO, HTTP, SSE)
Runs on ARM64/x64, edge to cloud
Cross-NAT device connectivity via Tai Link
Agent marketplace for one-command installs
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
