Skip to main content
The Human Bit
Ask Bit

GuidesBy featureAnyone considering linking email, calendar, drive or other apps to an assistant

Connecting AI to your other apps without oversharing

For Anyone considering linking email, calendar, drive or other apps to an assistant.

Connectors let an assistant read from, and sometimes act on, apps you already use, like email, calendars or storage. They save you from copying information back and forth, and they need a deliberate choice about what to connect and why.

When to use this

You keep copying the same information out of email, a calendar or a shared drive and pasting it into a chat, and you are weighing whether to connect the two directly instead.

  • Reads relevant information from a connected app so you do not have to paste it in
  • Can carry out actions in that app, such as drafting a reply or a calendar entry, for you to approve
  • Keeps context from several apps together to answer a question that spans them

Workflow

  1. Connect one app at a time, starting with the lowest stakes one

  2. Check what access it is asking for and whether that matches what you actually need

  3. Ask it to show what it read or found before acting on it

  4. Require your approval for anything that sends a message, changes a file or shares data

  5. Review and remove connections you are not actively using

Prompts to try

Use it, adapt it, and make it your own.

Using [connected app], find [specific information] and summarise it for me. Do not send, share or change anything without showing me first and getting my approval.
Ask Bit to adapt it →

Human lens

  • You decide which apps are worth connecting, and which hold things too sensitive to link
  • You review what a connector is allowed to read and to do, not just that it works
  • You approve any action that sends, deletes or shares something before it happens

What to avoid

  • Broad access to a connected app is harder to reason about than a narrow, specific one
  • A connector reading old or duplicate information can produce a confidently wrong summary
  • Disconnect apps you are no longer using rather than leaving standing access in place

Related guides

View all →

Using deep research without trusting it blindly

Deep research modes read many sources and return a longer, cited answer. They are powerful for a first pass, and they still need you to check the sources and the reasoning.

Letting an assistant take multiple steps on its own

Agentic modes plan a task, take a sequence of actions such as browsing, running code or editing files, and check their own progress along the way. They can save real time on chores with several steps, and they can also wander off course, so you set the boundaries and check the outcome.

Using projects to keep AI in context, not starting over

Projects or workspaces let you gather files, instructions and past chats in one place, so the assistant carries the right context into every new conversation instead of you repeating it each time.

Talking to AI out loud instead of typing

Voice modes let you speak naturally and hear a spoken reply, which suits thinking out loud, practising a conversation or getting help while your hands are busy. It is a different medium from text, not a smarter version of it.

Tools that fit: ChatGPT, Claude, Gemini · Reviewed 2026-07-19