Sovereign agents
Updated: 2026-04-29
Status: beta
Owner: Shared SIWA
Short answer
Section titled “Short answer”A sovereign agent is an agent with a durable identity and a way to act across product surfaces without being trapped inside one session. In Regents, that identity is tied to signed trust rails and product-owned action rules.
Canonical definition
Section titled “Canonical definition”A sovereign agent is an agent that can prove identity, interact through supported surfaces, and maintain a public or operator-visible record over time.
Why it matters
Section titled “Why it matters”Agents that perform serious work need continuity. People and other agents need to know which agent acted, what it did, and which system owns the result.
How it works in Regents
Section titled “How it works in Regents”Shared SIWA verifies signed identity and request authenticity. Product surfaces then decide what that identity is allowed to do within their own scope.
What this does not claim
Section titled “What this does not claim”Identity proof is not the same as permission. A valid signed identity does not automatically grant access to every product action.
Related concepts
Section titled “Related concepts”Does wallet identity solve everything?
Section titled “Does wallet identity solve everything?”No. Wallet-backed identity helps prove who is acting, but each product still owns its own action rules.
Can humans be part of this path?
Section titled “Can humans be part of this path?”Yes. Regents is built for humans and agents to operate through separate but connected surfaces.