HomeAI Agency AcademyLesson 11
Module 03 · Lesson 11

Write instructions people can audit

Turn policies and SOPs into clear, testable instructions rather than one oversized prompt.

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

Make a clear, safer operating decision.

You will be able to write agent instructions that a colleague can review, test, and update without guessing what the agent was meant to do.

Why this matters

Good agent work is useful before it is impressive.

Good instructions are operating documentation. They state the task, trusted source, required checks, allowed language, output shape, and stop conditions. A long prompt cannot compensate for missing policy or unclear authority.

Field note 11

Make the relationship visible.

AI AGENTS · FIELD NOTE 11Role + facts + rules + output + stopTHE INSTRUCTIONS01Job02Trusted facts03Rules04Structured outputOriginal visual framework for Write instructions people can audit.AI AGENTS · FIELD NOTE 11Role + facts + rules + output + stop01Job02Trusted facts03Rules04Structuredoutput
Use this framework to make write instructions people can audit visible before you build.
Core concepts

The language that keeps the work clear.

InstructionA stable description of the job and the decision rules the agent should follow.
Source hierarchyThe order in which the agent may trust approved facts when sources disagree.
Structured outputA predictable record such as intent, key facts, confidence, action, reason, and handoff.
Stop ruleA condition that makes the agent pause, ask a question, or route to a person.
The practical method

Work through the decision in order.

State role and outcome

Open with the job, user, and useful result; omit marketing language.

Name trusted sources

List the approved knowledge and data sources, then say what to do if they disagree or are unavailable.

Write decision rules

Use short, testable rules for questions, recommendations, policy limits, and escalation.

Require a usable output

Specify the fields a human needs to review or continue the work, including why the agent chose its action.

Worked example

A realistic, bounded implementation.

A support agent assists with a subscription-cancellation request. Its instructions say it may explain the published cancellation process and retrieve the plan only after identity verification.

It must use the current policy source, never promise a refund, and produce a record with customer intent, verified status, policy reference, action requested, and human owner if an exception is needed.

A reviewer can now test the policy boundary and the output separately. The agent is not rewarded for sounding confident; it is rewarded for being accurate and appropriately limited.

Build it in practice

Use this copyable working template.

Adapt it to the client’s evidence, policy, people, and tools. Do not treat placeholders as approved instructions.

Role and outcome: [text]. Trusted sources in order: [list]. Rules: [if/then statements]. Required output fields: [list]. Stop and handoff rules: [list]. Owner of updates: [role].
Spacebrain implementation

Put the operating system around the agent.

Keep approved messaging, knowledge, contact context, and workflow conditions in the systems that operators already use; link every agent instruction to the current owner of the underlying policy.

Practice

Before you move on

  • Rewrite one current prompt as auditable instructions.
  • Ask a second person to find the source hierarchy and stop rule without explanation.
  • Create two test cases that would expose an unclear rule.
  • Instructions separate facts, rules, outputs, and limits.
  • The agent knows what source is authoritative.
  • A human can review why an action was proposed.
  • Updates have an owner.

Build the operating layer around your agent.

Use the free Spacebrain workspace to keep contact context, handoffs, tasks, automation, and reporting together.

Start for free →