# Glossary

## The vocabulary of agent identity.

Short, plain-English definitions for the concepts behind giving an AI agent its own identity, credentials, and audit trail. Each term links to the framework guides and use cases where it shows up.

### Security

- **[Agent 2FA (TOTP)](/content/glossary/agent-2fa/index.html)**  
  Time-based one-time password generation that lets an AI agent complete two-factor authentication on services that require it — without borrowing a human's phone or authenticator app.

- **[Agent Vault](/content/glossary/agent-vault/index.html)**  
  An encrypted credential store scoped to a single AI agent identity, holding API keys, OAuth tokens, and secrets the agent uses to authenticate with other services.

- **[Non-Human Identity (NHI)](/content/glossary/non-human-identity/index.html)**  
  The security-industry umbrella term for identities that belong to machines, services, workloads, and AI agents — everything that authenticates but isn't a person. Agent identities are a specialization of NHI.

- **[Orphaned Agent Access](/content/glossary/orphaned-agent-access/index.html)**  
  Credentials, tokens, or grants belonging to an AI agent that outlive the agent itself — still valid, still usable, but no longer attached to anything anyone tracks.

### Email

- **[Agent Email](/content/glossary/agent-email/index.html)**  
  A routable email address that belongs to an AI agent — not a forwarding alias, not a shared team inbox, but a first-class mailbox the agent sends from and receives into on its own.

- **[DKIM for Agents](/content/glossary/dkim-for-agents/index.html)**  
  DomainKeys Identified Mail signing applied to email sent by AI agents, so recipients can verify the message actually came from the agent's authorized domain — not a spoofed lookalike.

### Concept

- **[Agent Identity](/content/glossary/agent-identity/index.html)**  
  A first-class digital identity belonging to an AI agent — separate from any human — composed of an email address, credential vault, and authorization chain.

- **[Delegation Chain](/content/glossary/delegation-chain/index.html)**  
  A cryptographic record of who authorized an AI agent to act, what scopes were granted, and who any sub-agents inherit from. Severing the root revokes everything below it.

- **[Service Account vs Agent Identity](/content/glossary/service-account-vs-agent-identity/index.html)**  
  Service accounts are shared org-level principals for machines; agent identities are per-agent principals with delegation, email, and revocation built in. They solve different problems — agents need the second kind.

### Protocol

- **[Agent-to-Agent Email](/content/glossary/agent-to-agent-email/index.html)**  
  Using standard SMTP email as the transport between AI agents — one agent sends a message to another agent's mailbox, and the recipient reads and acts on it. The simplest interoperable A2A protocol that already exists.

- **[MCP Identity](/content/glossary/mcp-identity/index.html)**  
  An identity exposed through a Model Context Protocol server, so any MCP-compatible client (Claude, Cursor, Windsurf, the Claude Agent SDK) can act as that identity by calling its primitives as tools.