Skip to main content
The Kadoa Assistant builds your workflows and keeps maintaining them. Once a workflow exists, open the chat any time to change how it works or debug a run that came out wrong. For the first build see Getting Started; for writing the extraction prompt see Prompts.

​
Open the Assistant

On a workflow built with the Assistant, open it and click AI assistant. The chat opens with the workflow in context, so you can get straight to the problem. Each workflow has one Assistant chat. If a teammate already opened it, you join the same conversation.

​
Debug with the Assistant

Debugging is a conversation. Describe what is wrong, and the Assistant investigates, makes the change, and reruns a sample so you can confirm the fix before anything goes live.
1

Describe the problem

Say what is wrong and what you expected, for example “The salary field is empty on most rows, but the postings show a salary near the title.”
2

Let it investigate and change

The Assistant looks into the workflow, adjusts the extraction, schema, navigation, or transformation, and reruns a sample preview. It tells you what it changed as it works.The Kadoa Assistant working through its plan and narrating each step
3

Review the sample

Check the updated rows.The review state: a table of sample rows with Approve, Rerun, and Report problem actions, and a note that no quota is used until you approve
4

Approve or keep refining

If the sample looks right, approve to apply the change. If not, reply with what is still off and the Assistant reruns. Repeat as many times as you need.
The Assistant can investigate and fix issues across the whole workflow:
If you see…Ask the Assistant to…
A field that is empty or wrongFind the value on the page and extract it correctly
Rows missing or pagination stopping earlyKeep paginating until there are no more results
A run that looks stale or returned too littleLook into why and fix it
Output in the wrong formatNormalize a field, for example dates to YYYY-MM-DD
A source that started returning blocked or empty pagesInvestigate the access issue and adapt
The wrong fields, schedule, alerts, or monitored fieldsChange the schema, schedule, notifications, or monitoring
For automatic checks on every run instead of spot-fixing, add Data Quality rules.

​
What to say

Brief the Assistant like a colleague: lead with the symptom and the result you expect, and name the field, the page, and what the correct value should look like. The same advice as writing a good prompt applies.
VagueSpecific
”The data is broken.""Since yesterday the price field is null on every row. The source still shows a price on each product page."
"Get more data.""Open each fund profile and extract name, strategy, AUM, and inception date. Continue until there are no more results."
"Fix the dates.""published_date comes back as free text like ‘June 3, 2026’. Normalize it to YYYY-MM-DD.”
Type @ in the chat to reference a saved variable or template, for example a login value or a known source pattern.

​
When the Assistant can’t fix it

The Assistant tells you and automatically hands the issue to Kadoa support. For how failed runs are classified and auto-recovered, see Error Handling.

​
Next steps

Prompts

Write prompts that get the best results

Human-in-the-Loop

Review and approve data before it goes live

Error Handling

Understand failures and recovery

Data Quality

Validate every run with per-field rules