Suggest actions: find the next step in a messy page
Ask AI to read a messy note, list, mind map, capture, meeting or event, and it suggests the tasks, habits, events and projects hiding inside. An AI to-do list that asks first: nothing is added until you say so.
What you can do with Suggest actions
- Ask from wherever the thinking is. "Suggest actions" is in the ⋮ menu of any note, list or mind map, on every capture, on meeting notes, and on each event's page.
- Tasks, habits, events and projects. Each suggestion shows the line from your page that prompted it, with dates, times or how often already filled in.
- Nothing is added until you say so.
- Tasks: add in one click.
- Events and habits: open ready-filled, so you can adjust them first; cancel, and the suggestion waits for you.
- Projects: open as a list of steps to edit, and nothing is created until you confirm.
- It asks when it isn't sure. A suggestion missing something important, like an event's date, asks you to review it instead of guessing.
- Linked back. What you add keeps a link to the page it came from.
- Choose how many. Ask for 3, 5, 7 or 10 suggestions, and refresh for a genuinely different set.
- Only when you ask. Nothing is read in the background, and a monthly allowance keeps AI costs predictable.
- Worth knowing: Suggest actions is part of Xale Pro.
Why I wanted an AI to-do list that asks first
When notes and plans pile up, finding the next useful action gets harder.
The goal of Xale is to have control, and in this rare chaos of too much going on, I wanted AI to do the noticing: read a long, messy page and point out what might need doing.
Your own desktop Agent can probably do this too, but we've built in simple functionality and an AI budget for Pro users to quickly pull out actions from their memories and find tasks and habits to work on.
Rachit, founder of Xale
The ideas behind Suggest actions
The model notices, you decide
AI is good at spotting what might be an action. Whether it matters is yours to judge.
- Worth reading: Hannah Fry, Hello World (2018). The best results come when people and algorithms work together, with the person keeping the final say.
- The usual way: rereading a long page with a highlighter, or an assistant that adds things to your list without asking.
- In this release: every suggestion waits for you, and the uncertain ones ask to be reviewed.
Only when you ask
A tool that reads your thinking should do it in the open.
- Worth reading: Ethan Mollick, Co-Intelligence (2024). Stay the human in the loop: use AI deliberately, and check what it gives you.
- In this release: AI runs on a click, on the page in front of you, and nowhere else.
Further reading
- Raja Parasuraman and Victor Riley, "Humans and Automation: Use, Misuse, Disuse, Abuse", Human Factors (1997), on how people come to over-trust automated suggestions. doi:10.1518/001872097778543886
Free to start. Already using Xale? sign in