Human-to-agent attribution

Trace a task back to a person without pretending that person personally executed every agent step.

01 / The operating case

What changes in a real workflow?

An analyst starts an incident agent that later disables an account.

Trace the authority boundary
  1. 01Owner
  2. 02Agent + workload
  3. 03Task context
  4. 04Action decision

02 / Decision contract

What the executor must check

Bind task issuance to the authenticated analyst and applicable delegation; retain distinct agent and approver identities at execution.

InputPrincipal · task · action · target
DecisionGrant ancestry · policy · current state
ResultEnforce · record · verify outcome

03 / Failure and evidence

What goes wrong, and what can be proven?

Failure mode

A log collapses human request, agent choice and human approval into one actor.

Evidence to retain

Task creation, originating user, agent identity, approval and account state.

04 / Canonical scope

Why this reference stands alone

Agent-to-tool attribution owns the downstream hop; this page owns originator accountability.

Technical artifact

Three decisions, three actors

Conceptual trace
analyst U7 creates incident task T
agent A proposes disable(account-42)
manager U9 approves request hash H
executor E applies H
originator U7 is not the approver U9

Run the denial test

Have U7 create the task and then spoof U9 in agent output. Only an independently authenticated approval event can satisfy the approval policy.

Primary references

Read the underlying material

Architecture discussion

Choose one consequential action and make its boundary explicit.

Request a Conversation