Skip to main content
When you use Clarity to document a process, you get a complete picture of how work happens today and where automation could help. This guide shows how to take that Clarity output and turn it into a working Duvo Agent — covering the decisions to make at each stage and an end-to-end example from start to finish.

Before You Start

Make sure these are in place:
  • A Clarity process with a generated map (Ready for Review or Completed) built from at least one analyzed capture
  • Team settings configured in Settings > Clarity (Company name, Industry, Average Hourly Rate) so Projected Impact figures are meaningful
  • A rough sense of which part of the process you want to automate first — you do not need a fully scoped plan, but a starting hypothesis helps

Step 1: Review the Clarity Output

Open your process in Clarity and review the Current Process and Automation Proposal tabs. Each gives you something specific you will use when building your Agent.

Current Process

Start on the Current Process tab to confirm how the team works today:
  • Steps — step-by-step breakdown of each action: who does it (Role), what system they use, inputs, outputs, decision conditions, and per-step Exceptions (including how the team handles each failure mode and how often it occurs). Mark steps that are:
    • Good automation candidates: repetitive actions, data lookups, form submissions, rule-based decisions, and anything handled the same way every time the process runs
    • Keep human for now: judgment calls based on context the Agent cannot see, sensitive communications, or actions where mistakes are hard to reverse
  • Summary — concise overview of the current workflow, roles, and systems.
  • SWOT — strengths, weaknesses, opportunities, and threats surfaced from your captures. Keep the weaknesses and per-step Exceptions open when writing your AOP (Agent Operating Procedure) — every failure mode listed there is a case your Agent needs to handle or escalate for human review.

Automation Proposal

Switch to the Automation Proposal tab to review how Clarity proposes transforming the workflow:
  • Transformation appetite and goal guidance — steer how bold the proposal should be (Transformational, Conservative, Optimize, or None) and add specific instructions before generating or regenerating the proposal.
  • Steps — the proposed future-state flow, showing which steps Duvo handles, which stay human, per-step automation readiness, and any flagged gaps that need more information before automating.
  • Summary — overview of the proposed automated workflow.
  • Projected Impact — estimated time saved, cost avoided, and business return with supporting assumptions. Use this to confirm you are automating the right slice first; if the figures look off, check your team Clarity settings or verify that all key people and systems were captured.

Building an Automation from the proposal

When you are ready to build from the full Automation Proposal, make sure any pending draft edits on the Automation Proposal tab are accepted or discarded first, then open the Automation tab on the process:

Prepare automation

Click Prepare automation. If the process is still in Ready for Review, Duvo asks Mark process as complete? — click Mark as Complete & Continue to mark it complete and start generating a draft automation plan (Agents, Queues, Connections, and handovers).

Review and create

Review and refine the draft automation plan, then click Create setup and Confirm to approve it and create the automation.
You can also build a narrower first Agent manually in the Agent Builder (see Step 3 below).

Step 2: Decide What to Automate First

Start with the smallest slice that delivers meaningful value. Trying to automate an entire end-to-end process in the first version makes testing harder and refinement slower. A good first version:
  • Covers three to seven consecutive steps from the Current Process / Automation Proposal
  • Has a clear start trigger (a file arrives, a Slack message comes in, a schedule fires)
  • Produces a clear, verifiable output (a row in a spreadsheet, a message in Slack, an email sent)
  • Handles the two or three most common failure modes documented in the step Exceptions and SWOT
Leave for later:
  • Steps that require significant judgment or context from outside the process
  • Edge cases that happen less than once a month
  • Connections to systems you do not yet have access to

Step 3: Build the Agent

If you want to build directly from the full Automation Proposal, use the Automation tab on the Clarity process. To build a focused first slice by hand, use the Agent Builder:

Open the Agent Builder

  1. Click + Create Agent from your dashboard.
  2. Select Use Agent Builder.

Write the AOP

Use the Steps panel (including per-step Exceptions), the Automation Proposal summary, and your own scoping notes to write the AOP. Describe the workflow as you would explain it to a colleague, starting with what triggers the Agent and ending with the final output.A practical AOP for a Clarity-derived Agent includes:
  1. The trigger — what starts the Agent (a schedule, a Slack mention, a file drop, a status change)
  2. The input — what data or context the Agent reads at the start
  3. The steps — the sequence of actions from the Steps panel, phrased as instructions
  4. Decision logic — the gateway conditions from the diagram that determine which path to take
  5. Exception handling — what to do when the situations documented in step Exceptions and SWOT occur
  6. Approval gates — where a human should review before the Agent proceeds, based on the high-risk steps you identified
  7. The output — what the Agent produces or updates when the Run is complete
Click Generate to create the AOP from your description. Review the output, then refine the description and regenerate if it does not match the workflow.

Add Human-in-the-Loop Approval Gates

For any step where a mistake would be hard to reverse — sending a customer email, updating a financial record, submitting a form to an external system — add a Human-in-the-Loop step in the AOP:
See Designing Human-in-the-Loop Workflows for guidance on where to place approval gates and how to phrase them.

Add Connections

Under Connections, enable the Connections to the systems involved in the process. Check the Steps panel in Clarity (or switch the diagram to Swimlanes grouped by System) to identify every system that appears — each one needs an authorized Connection on the Agent.If a system you need is not in the available Connections list, see Building Custom Connections.

Test

Click Start Work to run the Agent against a real or representative example. Watch the live execution view to see exactly what the Agent does at each step. Compare what you observe to the selected Automation Proposal steps and human handovers — or to the Current Process Steps panel if you intentionally scoped the Agent to reproduce the current flow.If the result diverges from your intended flow, return to the Agent Builder, note what differed, and update the AOP. See Refining Your Agent for a structured approach to iteration.

End-to-End Example: Supplier Invoice Follow-Up

The situation: Your procurement team spends about an hour each week chasing suppliers for overdue invoices. The process involves checking a spreadsheet for invoices past their due date, looking up the supplier’s contact email in the ERP, drafting a follow-up email, getting a manager’s approval, and sending the email. Three people are involved but the work is largely the same each time. What Clarity found:
  • The Current Process Steps panel showed seven steps, five of which were the same every time regardless of supplier or invoice amount
  • Step Exceptions and SWOT flagged invoices over $50,000 and suppliers with an active dispute as cases that needed extra care
  • The Automation Proposal Projected Impact tab estimated 3.2 hours saved per week at current team size
  • The Automation Proposal proposed reading the overdue invoice list, drafting follow-up emails using templates keyed to invoice age, and routing anything above $50,000 or flagged as disputed to the procurement manager for approval before sending
The scoping decision: Automate the five repeatable steps. Keep manager approval for high-value and disputed invoices, and start with invoices under $50,000 to de-risk the first version. The AOP:
Connections added:
  • Google Sheets (to read and update the Procurement Tracker)
  • [ERP system via Custom MCP] (to look up supplier contact emails)
  • Gmail (to send follow-up emails)
  • Slack (to post the summary)
Trigger: Weekly schedule, Monday at 9:00 AM First test result: The Agent correctly identified four overdue invoices, drafted professional follow-up emails, and escalated one invoice above $50,000 to the procurement manager. One email had the wrong contact email because the ERP lookup returned a primary contact rather than the AP contact — the AOP was refined to specify “accounts payable contact” and the test was rerun successfully. What expanded later: After two weeks of reliable operation, the team added the disputed invoices workflow (a different email template and a different escalation path) and extended the schedule to run twice a week.

Tips for Better Results

Use your target Steps panel as a checklist: After the first Run, compare what the Agent did against the selected Automation Proposal steps and human handovers (or the Current Process Steps panel when reproducing the current flow) step by step. Any unintended divergence is a specific refinement to make in the AOP.
Copy exception handling verbatim from step Exceptions: The exceptions in Clarity are grounded in what your subject matter experts described, along with how the team handles them. Copy that language directly into the AOP rather than paraphrasing — it preserves the nuance.
Tune the Automation Proposal before building: Before building an automation from Clarity or copying steps into an AOP, adjust the Transformation appetite, update goal guidance, or use Edit mode / Ask Duvo on the Automation Proposal to keep sensitive steps human and specify which systems to connect.
One scope, one Agent: If the Clarity process covers multiple distinct workflows — for example, both routine invoices and disputed invoices — create separate Agents for each rather than a single Agent with complex branching logic. This makes testing and refinement easier.

Clarity

How to capture processes and generate documentation

Agent Builder

How to describe and generate AOPs

Designing Human-in-the-Loop Workflows

Placing approval gates correctly

Refining Your Agent

Iterating after the first test Run

Test Safely Before Going Live

How to test without affecting production systems