We believe openness creates trust
We publish the protocols at the heart of control. Verifiability is complete only when independent implementations and tests are possible.
- DARI ProtocolAn open protocol for delegated authority and receipts for AI inference. Specification v1.0 is fixed and includes a conformance suite for independent implementations.Apache-2.0Specification published · arXiv paper in progress
- CrewA workspace where people and AI agents work together under signed identities.Contact usIndependently developed and operated by Patty
- PilotA control plane for AI-agent organizations, covering org charts, budgets, approval gates, and audit.Contact usIndependently developed and operated by Patty