Working with the agent¶
Questo has one agent. The same agent answers you in chat, runs your tasks in the background, works through the routines and replies in Slack, and it reads the same knowledge base every time.
Where the agent is¶
The chat column sits beside every page except Settings. The Start page in the sidebar is the agent on its own: a message box, a Quick action menu (new mission, new task, new skill, new rule, update knowledge), and a view of the day with what needs action in your inbox, what is ready in your Outbox and what is up next.
Chats are tabs¶
The chat is not one endless conversation.
Each tab has its own memory. Two tabs open on two different problems will not bleed into each other. This is what makes it safe to leave a long thread running while you ask something unrelated.
New chat is genuinely fresh. Not a visual clear with the old conversation still in scope. If a thread has gone somewhere unhelpful and you cannot steer it back, starting a new chat actually works.
All chats lists your active and archived tabs and searches them. Archived chats are read-only.
Type @ in the message box to point the agent at something in your workspace. While the agent is working, Stop interrupts it, and anything you send in the meantime is queued as a follow-up. A run that has gone the wrong way does not have to finish before you correct it.
It sees what you have open¶
When you are looking at a mission, task, output, insight or your knowledge base, the agent knows which one. You can say "add a task for this" without naming it, and when it edits something on your screen you watch the change land.
Every change can be undone¶
The agent applies its work directly instead of asking first. After each turn, a receipt lists what it changed, with Undo on each change and Undo all for the turn. History on outputs, missions, tenets, records and the knowledge page keeps every change from every source, yours included, and Undo this run reverses everything one background run did.
What the agent will and will not do on its own is in what the agent does without asking.
How to frame a prompt¶
The most common mistake is writing a specification. The agent has your knowledge base; it does not need you to restate your positioning in the prompt.
The second most common mistake is being so brief that "good" is undefined. "Write a blog post" gives the agent nothing to aim at.
A prompt worth writing says:
- What you want, in one line.
- Who it is for, when it is not obvious from the mission.
- What good looks like, which is the part people skip. "Something a sceptical CFO would forward" tells the agent more than three paragraphs of structure.
- What to avoid, if you have been burned before. Though if you find yourself writing the same avoidance twice, it belongs in a rule.
A good first prompt is usually three or four sentences. If yours is running to a page, most of it is probably context that belongs in the knowledge base, where every future task can use it too.
Personal preferences¶
Facts about you that the agent should always know, such as your role, your LinkedIn profile or how you like to be addressed, go in Settings > Preferences. Only you see them, and the agent can add to them when you tell it something about yourself in a chat.
From Slack¶
With Slack Agent connected, mention @Questo in a channel it has been invited to and the agent answers there, with your workspace's full context. It works as whoever mentioned it, matched by the email on their Slack profile, so only members of your Questo workspace can use it. See Integrations.
From your coding tools¶
Claude Code, Cursor and other MCP clients can read and update your workspace as you. See MCP access.