Skip to content

Learning Space

Learning Space is DeepTutor’s library and personalization layer. It is where the things that persist live — your past conversations, the materials you’ve saved, and the presets that shape how the assistant behaves. Everything here can be reused later from Chat, Partners, Co-Writer, and Book.

Open Learning Space in the left sidebar. The hub is organized into two groups — Conversations & Materials and Personalization — each tile opening a full manager.

The Learning Space hub: Chat History, Notebooks, Question Bank, Mastery Path, Personas, Skills

Every conversation you have is retained here. Browse, search, rename, delete, and reopen any session to keep going — and reference one in a new turn via the composer’s + menu.

Chat History: searchable list of past conversations

Notebooks are the easiest way to turn an answer into durable, reusable context. Create a notebook, then save records into it from Chat, Research, Partners, or Co-Writer — each record keeps its source and timestamp. Attach a notebook later from Chat or compile one into a Book.

Notebooks: create a notebook and inspect its saved records

Quiz questions you generate are collected here with your answer, the reference answer, whether you got it right, and an explanation. Bookmark items, group them into categories, filter by Wrong Only, and jump back to the original chat to review in context.

Question Bank: a saved question with your answer, the reference answer, and an explanation

A Mastery Path is a structured learning plan made of objectives, each typed as a concept or a procedure and tracked through mastery states. It enforces hard gating (you advance only when an objective is mastered) and spaced review (objectives come back when they’re due). You study a path by continuing it in Chat — it runs on the Mastery Path capability.

A Mastery Path: objectives with concept/procedure tags and per-objective mastery state

Personas are behavior presets that shape the assistant’s tone and stance, applied per turn from the composer. The built-ins — peer, research-assistant, teacher — are good starting points; create your own with New persona, and reuse any of them when creating a Partner.

Personas: peer, research-assistant, and teacher presets

Skills are SKILL.md playbooks the model reads on demand — they teach DeepTutor a workflow, style, or domain behavior. A good skill is specific: it says when to use it, what inputs it expects, and what output is useful. Skills are tagged and searchable; built-in office skills (docx, pdf, pptx, xlsx) and presets are read-only, while skill-creator helps you author new ones.

Skills: tagged capability playbooks, including built-in office skills

You don’t have to write every skill yourself. Import from EduHub browses the community catalog and downloads a skill straight into your library — through a security gate (hub verdict check, safe extraction, no binaries, no prompt self-injection).

Import skills from EduHub: browse and download community skills

The same flow exists on the CLI:

Terminal window
deeptutor skill install clawhub:<slug>
  • Chat Workspace — attach Learning Space assets to a turn
  • Partners — assign skills, notebooks, and personas to a companion
  • Interactive Book — compile notebooks and question banks into a book
  • CLI commands — manage skills and assets from the terminal