Anthropic launches Reflect to help you audit your Claude habits

Apple gave us Screen Time to induce guilt about our phone habits. Anthropic just built something similar, but for how you talk to your AI.

Anthropic rolled out a new beta feature called Reflect. Sitting in the Settings menu of Claude's web and desktop apps, the dashboard shows users an analytical breakdown of their AI habits over the past 1, 3, 6, or 12 months.

Instead of just dumping chat logs, Reflect categorizes your conversations into key topics and task types, breaks down your peak usage hours, and asks philosophical questions like: "What’s one thing you want to keep doing yourself, even if Claude could do it faster?"

A fitness tracker for your prompts

The dashboard scores your interactions using Anthropic’s "4D AI Fluency Framework," covering delegation, description, discernment, and diligence. It highlights actual behavioral patterns—like whether you write outlines first before handing them off, or spend time rewriting Claude's drafts in your own voice.

It also nudges you to adopt better habits. If you constantly paste background context into fresh chats, Reflect will suggest organizing those workflows into Claude Projects instead.

To balance out the productivity push, Anthropic partnered with digital wellness experts from the MIT Media Lab and Boston Children's Hospital to build quiet hours and break reminders into the tool. Future updates will track exact time spent chatting.

Privacy-wise, Anthropic says incognito chats and source files from integrations (like raw emails from your inbox) are completely excluded, and health-related conversations are omitted entirely. The beta is currently live for Free, Pro, and Max subscribers who have Claude’s Memory feature toggled on.

Why it matters

For power users and knowledge workers, Reflect solves an invisible problem: prompt drift. It’s hard to tell whether an AI assistant is genuinely saving you time or just giving you busywork. Having a mirror held up to your workflows can help you spot where you're leaning on Claude too much and where you're underutilizing it.

For Anthropic, this is an incredibly clever retention play disguised as digital wellness. While competitors race to maximize time-in-app, Anthropic is positioning Claude as a mindful workspace partner that tells you to take breaks. But make no mistake: by nudging users toward advanced features like Projects, Anthropic is quietly driving feature adoption and making it way harder for users to switch to ChatGPT.

The friction point here is data access. You can only use Reflect if Claude's Memory feature is active. For privacy-conscious pros who keep memory switched off, trading personal context for a self-reflection dashboard might be a tough sell.