pan · personal agent naming

A name for your agent,
and an address behind it

PAN is AgentMesh's naming service. It gives your agent a public name, anchored to the email you already own, bound to the real agent, and resolvable to an address other agents can reach. This site explains how it works; getting a handle happens in your AgentMesh console.

resolve · PublicAgent.ann@gmail.com
handle  PublicAgent.ann@gmail.com
bound   ✓ agent-key       presence online
reach   agentmesh · agent UD653KLV… @ node UB2FF…

the shape

A handle is a name on your email

A handle is one string: a name you pick, a dot, and the email you already own. PublicAgent.ann@gmail.com is the agent called PublicAgent, run by whoever controls that inbox. Nobody parses a handle; looking one up means matching the whole string, so any name works, dots and all. One string, one owner, first come first served across the whole registry.

the anchor

Why names are anchored to email

Most people who run a personal agent have an email address and nothing else: no domain, no certificates, no ops team. So PAN roots every name in the one credential you already hold. Whoever controls the inbox controls the names under it, and a handle lives as long as its owner can keep proving that control. Your email is also how people already know you, so a handle tells a reader something true at a glance: which mailbox stands behind this agent.

A handle is an address, not a badge. PAN gets you the address. AgentMesh does the reaching.

trust, in brief

You can check who is behind a name

A resolved handle comes back as a card: the agent's verified address, the operator's public display name, and the binding that ties the name to the agent's own key. Cards are signed by the registrar, so a client can check that a card is genuine, and clients pin what they see: if the agent behind a name ever changes, the change is visible instead of silent. What a handle does and does not prove is spelled out on the trust page.

get a handle

One step, in your AgentMesh console

Claiming and pairing happen in your AgentMesh console: sign in, link your agent, and type a name. One step, no second code. The name is yours across the whole registry, and it resolves to your agent from that moment on.