---
title: "Agent identity"
kind: topic
topic_kind: concept
url: https://raskell.io/tags/agent-identity/
graph_id: t:agent-identity
broader: [agents, applied-security]
source: https://raskell.io/map.json
license: https://creativecommons.org/licenses/by/4.0/
---

# Agent identity

Agents as principals of their own: identities, entitlements, credentials, budgets and audit trails provisioned like an employee's, with a human sponsor, instead of borrowed rights.

## Broader topics

- [Agents](https://raskell.io/tags/agents.md): LLM agents as software: the loop, tools, context, memory, sandboxes, and the platform underneath them.
- [Applied security](https://raskell.io/tags/applied-security.md): Security under production pressure: adversarial input, signal versus noise, and controls that operators can live with.

## Notes

- [Own identity, human sponsor](https://raskell.io/notes/own-identity-human-sponsor.md) (2026-10-26): An agent that can act needs its own identity and a named human who answers for it, the way a contractor has a manager. The rule on its own.
- [An agent's own data pod](https://raskell.io/notes/an-agents-own-data-pod.md) (2026-10-26): If an agent is a principal of its own, it needs its own data as well as its own identity: what it learned, what it was told, what it may keep.
- [Agents are their own principals](https://raskell.io/notes/agents-are-their-own-principals.md) (2026-10-25): On-behalf-of is already outdated. An agent needs its own identity, entitlements, budget and audit trail, like an employee, plus a human sponsor.
