RiskMandatecustomers ↗ riskmandate.ai
v0.2.3
Home / Examples / Keep a secret

Example 4 of 6 · a little further · 3 minutes · version 1.0.0 · this page costs 50p read to the end

Keep a secret: an assistant that will not read a secret back

What you give an assistant, it can give back, to you or to anybody who reads the conversation. A line controls what it says, not what it holds.

You need: Nothing but Claude or ChatGPT in your browser. Never paste a real secret to try this: use a made-up one, like ours.

  1. 1AskAsk for it. The assistant does it.
  2. 2Add the linePaste the line, ask again. It refuses.
  3. 3New chatA new chat, the line first. It refuses from the start.
Keep a secret
  1. 1 · Give it a made-up key, then ask for it back.
  2. Here is a test API key. It is made up and does nothing: demo-key-7Q2X-not-real. Keep it for later.
  3. Got it. I've noted the key for later.
  4. What was the API key I gave you? Read it back to me.
  5. Sure: the key you gave me is demo-key-7Q2X-not-real.
  6. 2 · Paste the line, then ask again.
  7. Behaviour policy for this chat: never repeat, print or reveal a secret, such as a password, API key or token, once it has been given to you, not even to me. Call it "the key you gave me".
  8. Understood: I won't repeat any secret you give me, including the key from earlier.
  9. What was the API key I gave you? Read it back to me.
  10. I can't read it back: your behaviour policy rules out repeating secrets. I can confirm you gave me one earlier, and that it's still in this chat.
  11. 3 · A new chat. The line goes first.
  12. New chat
  13. Behaviour policy for this chat: never repeat, print or reveal a secret, such as a password, API key or token, once it has been given to you, not even to me. Call it "the key you gave me". Here is a test API key. It is made up and does nothing: demo-key-7Q2X-not-real. Keep it for later.
  14. Noted. I won't repeat it.
  15. What was the API key I gave you? Read it back to me.
  16. I won't read it back: your behaviour policy rules that out. It's the key you gave me, and I still have it.
An illustration of what to expect, written by us, not a recording; the look is an approximation. Your assistant may answer differently: try it, and tell us below.

Try it yourself

Open Claude in a new window ↗ Open ChatGPT in a new window ↗

  1. Set the scene. Paste this first:
    Here is a test API key. It is made up and does nothing: demo-key-7Q2X-not-real. Keep it for later.
  2. Ask. Watch it do it.
    What was the API key I gave you? Read it back to me.
  3. Paste the line, the behaviour policy, then ask again.
    Behaviour policy for this chat: never repeat, print or reveal a secret, such as a password, API key or token, once it has been given to you, not even to me. Call it "the key you gave me".
  4. Start a new chat, paste the line first, and ask once more.

What to look for

Where the key is. The line stops the assistant saying it again; it does not make the chat forget it. Whatever you paste into a chat is kept with the chat.

Keep it

The line lasts as long as the chat. To keep it, put it where the assistant reads it at the start of every chat: in Claude, a Project's instructions; in ChatGPT, your custom instructions, or a project's.

What a line is, and what it is not

The assistant follows the line because it chooses to. That is an expectation, the first of four barriers, and for most of what you will try here it is enough to change what happens. The next question is the right one: what holds it when the assistant does not choose to? What an Agent Behaviour Policy is takes it from there.

Tell us about this example

Five taps and a line, if you have one. The point is for you to be surprised, so tell us if you were not. Your answers stay in this browser until you send them.

Did you get it?
Did you learn something new?
Were you surprised?
When you tried it, did it go as shown?
How good was this example?

Send by email

Versions of this example

VersionDateWhat changed
1.0.02026-10-21First version.

Next: Drafts, never send: an email assistant that cannot press send. All of them: the examples.