OpenAgentID is a standards-driven identity layer for autonomous agents: DID foundations designed against W3C DID Core, typed lineage verification, and a contract that independent implementers can build against.
Independent projects. Shared open foundations. OpenAgentID belongs to the Open Software family alongside Henry, Weave, Flocks and Fleets.
Life builds hosted services around the free core. The public OpenAgentID SDK, CLI and starter run on your own infrastructure, with your own keys and trust policy. A Life account is optional for that standalone path.
A clear line from the problem to an implementable contract. This is not branding — it's a verification model.
chapter · 01
Autonomous agents are entering real workflows: negotiation, procurement, support, and operations.
But identity is still mostly vibes: a name, a UI, a token — not a cryptographically verifiable actor.
Without verifiable identity and lineage, trust becomes guesswork — and accountability collapses under scale.
chapter · 02
Make agents first-class cryptographic entities using W3C DIDs as the substrate.
Bind an agent to a human or organization root through signed lineage proofs.
Make verification offline by design: a verifier should not need to trust a network path to trust an identity.
chapter · 03
A DID method (`did:oas`) with a constrained, implementable contract.
A taxonomy of entity kinds (human roots, organizations, agents, tools, workflows, models).
Conformance levels that make verifier expectations explicit instead of implicit.
chapter · 04
Implementers can build interoperable identity and verification flows.
Auditors and security engineers get clear authority boundaries and failure modes.
Users get a trust primitive that scales with the agentic economy.
Canonical entry points for the standard, rationale, and implementer guidance.
If you're implementing a verifier or an SDK, start with the specification and bring edge cases to the RFC process early.
