Features
Captures AI prompts, reasoning, and discussions
Links AI sessions to Git commits for traceability
Injects structured repository context (architecture, patterns, constraints)
Enforces architectural constraints on AI-generated code
Local-first: runs offline, data stored in repository
Works with Claude Code, Cursor, Codex, Gemini, Copilot, OpenCode
Semantic model built continuously via AST analysis
Commit-aware context retrieval to reduce token consumption
Faster onboarding for new team members
Open-source CLI tool (Apache 2.0)
Repository-scoped context
Supports constraint enforcement (coming soon)
Agent-agnostic: works with multiple agents simultaneously
Auto-detects and connects AI assistants via 'bitloops init'
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
