HomeAI Agency AcademyLesson 12
Module 03 · Lesson 12

Design tools, approvals, and handoffs

Give an agent only the actions and information the job truly needs.

Last updated August 5, 202615–25 minutesFree AI agent course
What you will learn

Build the tool and approval matrix

You will be able to set least-privilege tool access, approval boundaries, and a handoff that lets a person continue without repeating the customer’s story.

Why this matters

Good agent work is useful before it is impressive.

Most avoidable agent failures come from too much authority, not too little intelligence. The first safe version often reads information, drafts a recommendation, and asks a human to approve the consequential action.

Field note 12

Move from read to recommendation, approval, and action

Move from read to recommendation, approval, and action
Move from read to recommendation, approval, and action
Core concepts

The language that keeps the work clear.

Least privilegeAccess only to the data and tools required for the job—nothing broader.
Reversible actionA change that can be safely undone with a clear record.
Approval gateA human decision required before an agent sends, changes, charges, deletes, or commits a sensitive action.
Handoff packetThe short context a person needs: request, evidence, recommendation, reason, transcript, and next action.
The practical method

How to build the tool and approval matrix

List every tool

Write each lookup, write, send, schedule, payment, or deletion capability separately.

Rank consequences

Mark which actions are informational, reversible, customer-facing, financial, sensitive, or irreversible.

Set the first boundary

Start with read-only or recommend-only when uncertainty or impact is high.

Build the packet

Require the agent to preserve the facts, reason, customer request, and recommended next step for the human owner.

Worked example

Worked case: design tools, approvals, and handoffs

An agency builds an appointment assistant for a contractor. It can read service areas, appointment availability, and a lead’s stated needs. It can recommend a slot and create a draft booking record.

It cannot cancel an existing appointment, promise a price, or message a customer outside approved templates without an owner’s approval. If the lead describes a safety risk, it creates an urgent task and uses the approved emergency-routing message.

The dispatcher receives the full handoff packet, not a vague notification that says ‘AI needs help.’

Build it in practice

Complete the working artifact

Tool: [name]. Purpose: [why needed]. Access: [read / recommend / act]. Consequence if wrong: [impact]. Approval required: [yes/no + owner]. Handoff fields: [list].
Authorization check

Tool calling is an authorization boundary.

A valid function name and JSON shape do not make an action safe. The application must validate arguments, authenticate the acting identity, authorize the specific resource, limit the operation, and return an explicit result the agent can interpret.

  • Use typed inputs and reject unknown fields.
  • Make write operations idempotent where possible.
  • Require approval for actions whose cost or consequence cannot be reversed easily.
  • Log the request, identity, decision, result, and error without exposing secrets.

Sources used for this check

Practice

Before you move on

  • Inventory the proposed agent’s tools.
  • Downgrade one risky action to recommendation or approval.
  • Review whether the human can reconstruct what happened from the handoff packet.
  • Every tool has a job-specific purpose.
  • High-impact actions have approval or are removed.
  • The handoff contains evidence and a next step.
  • The first release is more limited than the eventual ambition.

Failure drill

A harmless tool call becomes a customer-facing mistake

The case

An agent may draft a refund, issue it, and email the customer through three tools. The same instruction covers all three actions, and the approval screen shows only a generic 'continue' button.

Your call

  1. Which actions need separate authority?
  2. What must the approver see before acting?
  3. How will the system prevent the same action from running twice?
Reveal a defensible response

Response: Separate drafting, approval, execution, and notification. Show the customer, amount, reason, policy source, and exact action at approval time. Give the write operation a narrow permission and an idempotency key; log who approved it and what happened.

Lesson progress

Finished this lesson?

Save your place on this device so it is easy to pick up where you left off.

Not marked complete yet.