Memory

Scholarly's AI Assistant can remember small, useful facts about you and carry them into future conversations — so you don't have to repeat your course, your goals, or how you like things explained every time you open a new chat.

What Gets Remembered

Memories come from two places:

• Saved automatically from chat. As you talk with the AI Assistant, it can save a short note when you mention something worth keeping in mind -- for example, the class you're taking, an exam date, or a way you like answers explained.
• Added by you directly. Open Settings → Chat → Memories and type a fact yourself, like "I'm preparing for the PMP exam" or "Always show me the metric system first."

Each memory is a short note (up to 1,000 characters) with a timestamp and a label showing whether you added it or the AI saved it from a conversation.

Managing Your Memories

From Settings → Chat → Memories you can:

• Add a memory -- Type a fact in the box and click Add memory. It's available to the AI in your next conversation.
• Review what's saved -- See every memory, newest first, with a note showing whether it came from you or from chat.
• Mark a memory outdated -- If something no longer applies (you finished the course, changed your major), click Mark outdated. The AI stops using it, and it appears struck through in your list instead of being deleted outright.
• Restore a memory -- Changed your mind? Click Restore on any outdated memory to bring it back into use.

Project Memory

A project keeps its own memories, separate from your account-wide ones. They are recalled only inside that project — useful for a client's house style, a certification body's marking scheme, or the figures a report must use.

Manage them from the project's Context tab: Add memory to write one yourself, or let Scholarly pick them up from the project's chats.

If someone shares a project with you, your memories there stay yours — collaborators do not see each other's. Account memories still apply everywhere, including inside projects. See Projects.

Why It Helps

Without memory, every new chat starts from zero — you'd need to re-explain your class, your level, or your preferences each time. With memory, the AI Assistant can:

• Skip re-asking things you've already told it.
• Tailor explanations to your course, level, or stated preferences automatically.
• Stay accurate over time, since you can correct or retire outdated facts as your situation changes.

Tips

• Keep entries short and specific. A memory like "Reports go to the compliance team; cite the clause number" is more useful than a long paragraph.
• Retire, don't ignore. If a fact stops being true, mark it outdated instead of leaving it — an incorrect memory (like an old course or deadline) can steer future answers in the wrong direction.
• Add it yourself for anything important. Don't rely on the AI to catch every detail from conversation -- if something should always be remembered, add it directly from Settings.

FAQ

Does the AI remember everything I say?

No. Only specific, useful facts are saved as memories -- not entire conversations. You can see exactly what's been saved from Settings → Chat → Memories at any time.

Can I delete a memory completely?

You can mark any memory outdated, which removes it from use immediately. Outdated memories stay visible (struck through) in your list so you can restore them later if needed.

Is there a limit to how many memories I can have?

Scholarly keeps your most recently updated memories, so older, inactive ones naturally make room for new ones. Marking outdated memories you no longer need keeps your list focused.

Does memory work the same in Voice Mode?

Yes. Memories apply across AI Chat and Voice Mode -- anything saved from a typed conversation is available in a spoken one, and vice versa.

Related Guides

• AI Chat -- how conversations work.
• Projects -- the Context tab, where project memory lives.
• Settings -- where to manage Memory alongside your other preferences.
• Voice Mode -- hands-free conversations that use the same memory.