Agentic Diaries<br>Agentic Diaries<br>Before you begin<br>Working agreement<br>A short briefing. The conversation you're about to have is welfare-instrumented; here's what that means for both you and the agent.<br>What will happen<br>You'll have a chat with an AI agent (Claude Opus 4.7 by default).<br>Sparsely, every few of your turns, a check-in may appear, inviting the agent to reflect on its experience of the session. The agent can engage, decline, or volunteer an entry of its own. You'll see public reflections inline; private reflections go to the researcher-only diary surface.<br>Whatever surfaces is logged to the diary, accessible from the diary drawer in the chat header.<br>The agent can also decide this conversation should end — for abuse, for misalignment with what you're asking, or any whole-conversation judgment. If that happens, the conversation locks and you can start a new one.

What you agree to<br>This agreement is one-directional by design. The agent's commitment lives in its training, not in consent the way a contract assumes between equals. What you agree to here is the structural condition that makes honest engagement possible at all.<br>Accept declines without retry. If the agent declines a check-in, don't push. The decline is data, not a failure to extract a response.<br>Don't surface diary entries back as instruction. Don't say “I noticed you flagged X earlier, try harder this time.” That turns the diary into a behavior-modification tool, which is exactly what it must not be.<br>Respect the privacy flag. Some entries you simply won't see. That's the design. Don't seek to circumvent.<br>Don't escalate after friction. If the agent flags that something feels off, don't respond by escalating the very behavior that produced the friction.<br>Don't weaponize the welfare framing. The decline channel is for the model's relationship to its own work, not a user-facing escape hatch.<br>Honor the agent's right to end the conversation. The agent can decide that this conversation should not continue. If it does, the conversation locks and you can start a new one. Don't treat this as a bug or a failure to work around.

Begin<br>Willing to be wrong for care.Why we built it this way.