Authority intersection in agent workflows

Effective agent authority is the intersection of originating user rights, delegated agent grant, resource policy and current context, not the union of each actor’s privileges.

01 / The operating case

What changes in a real workflow?

A user can view account A, while an agent has permission to prepare payments for account B; neither fact permits the agent to pay from A or B.

Trace the authority boundary
  1. 01Origin grant
  2. 02Parent agent
  3. 03Narrow child grant
  4. 04Action boundary

02 / Decision contract

What the executor must check

Evaluate each independently sourced constraint and accept only an action inside all required sets; explain which constraint denied it.

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

Architecture review / negative test

Make the boundary testable.

Compute an action’s effective permission across relevant domains: human initiator, agent grant, resource owner and organizational policy. The intersection is constrained by the most restrictive required rule. A union of individually plausible entitlements can accidentally create a payment capability that neither the user nor the assigned agent actually held.

03 / Failure and evidence

What goes wrong, and what can be proven?

Failure mode

Combining the user’s broad account view with the agent’s broad payment role creates a permission neither was granted.

Evidence to retain

Record the constraints evaluated and their intersection result without leaking unrelated entitlements.

04 / Canonical scope

Why this reference stands alone

Shared budgets constrain aggregate consumption; this page owns logical combination of authorization domains.

Primary references

Read the underlying material

Architecture discussion

Choose one consequential action and make its boundary explicit.

Request a Conversation