Workflows
A research workflow that ends with sources
A research run in resolvenengine happens in visible, grouped tabs, skips pages that add nothing new, and finishes with a source block, so every claim can be traced back to the page it came from.
Last updated 18 September 2026
Write the prompt like a brief
- 1State the question and the decision it feeds.
- 2Name the scope: how many sources, which market, how recent.
- 3Say what the output should be: a table, a shortlist, a summary per row.
- 4Attach a CSV or a link list when you already know where to look.
What happens during the run
- Tabs are opened in a labelled group so you can follow along.
- A ledger tracks which pages contributed something new.
- Repeated dead ends stop the run rather than looping.
- The answer separates what was observed from what was inferred.
Finish with a check
Open two or three of the listed sources before you act on a result. A research answer without a source block is not finished work, and the assistant is built to say when something could not be verified.
Frequently asked questions
Will it flood my browser with tabs?
No. Tabs are grouped, empty tabs are reused, and the group can be closed in one action.
Can it research my own data instead?
Yes, that goes through the connectors rather than the browser.
Related guides
Working in your current tab, not a random new oneHow resolvenengine prioritises the tab you are looking at, when it opens its own tabs, and how the visible ResolvEngine tab group keeps work separate.Attach files: CSV, spreadsheets, PDFs and imagesAttach up to ten files per message, preview tables, text and PDFs in a full-panel viewer, and use link lists from a CSV as task input.Choose a model and keep cost predictableWhy resolvenengine uses your own API key, how to pick a model for a task, and which controls keep token spend predictable.
Try it in your own Chrome
Pro is $19 USD per month and uses your own AI API key, so the provider bills model usage at their own rate.