§01Federation · prerelease

OpenID Federation 1.0for JavaScript.

The complete OpenID Federation 1.0 implementation for JavaScript — runtime-agnostic, spec-compliant, built on Web API standards.

Trust Anchors, Entity Configurations, Trust Chains, Subordinate Statements, Metadata Policy, Trust Marks, authority hints, federation endpoints, and automatic / explicit client registration — wired as (Request) => Promise<Response> handlers that run identically on Node.js, Deno, Bun, workerd, Electron, and browsers.

Apache 2.0 · MIT6 runtimes
resolve.trust.chainES256
OpenID Federation trust chain diagramTrust Anchorta.example.orgIntermediateia-a.example.comIntermediateia-b.example.netOPop-1.example.comRPrp-1.example.netOPop-2.example.orgRPrp-2.example.com
chain.validverified
statements3
policy_ops4
expires_in23h 47m

ta.example.org → op-1.example.com · resolved in 320ms

§02Why Federation

Trust, unbundled from bilateral agreements.

Traditional approaches to establishing trust between systems rely on bilateral agreements and manual metadata exchange. OpenID Federation introduces cryptographically verifiable trust chains — enabling dynamic, scalable trust without per-party configuration.

O(depth)
  1. No more bilateral agreements

    Entities join a federation once. Trust is derived from a cryptographically signed chain — not from individual contracts between every pair of participants.

  2. Verifiable trust at scale

    Every claim is signed. Trust Anchors publish constraints and metadata policies that are cryptographically enforced down the chain.

  3. Protocol-independent by design

    Works with OpenID Connect, OAuth 2.0, and beyond. The federation layer is orthogonal to the protocol used for authentication or authorization.

§03What's Inside

Four spec packages, three apps, one CLI.

Modular by design. Install only what you need — from core primitives to full OIDC registration flows, interactive learning, and visual exploration tools.

TypeScriptRuntime-agnosticWeb API
§05AI & Machine Identity

Verifiable trust for agent-to-agent futures.

As AI agents interact on behalf of users and organizations, verifiable trust becomes critical. OpenID Federation provides the infrastructure for agent-to-agent trust — machines can verify each other's identity and capabilities through the same cryptographic trust chains that today secure humans, applications, and services.

The same Trust Anchors, Entity Configurations, Subordinate Statements, Metadata Policy, and Trust Marks that bind humans into a federation can bind autonomous agents into a federation of machines — with cryptographically enforceable limits on what any given agent is authorised to do on whose behalf.

The OpenID Foundation's AI Identity Management Community Group has published a whitepaper (PDF) naming OpenID Federation as a candidate trust fabric for agent-to-agent identity.