madhvantyagi/Gnos

Teaching harness , help you to learn anything , It teaches like real teacher , design curriculum , generate videos , simulations , images , pdfs , tracks your learning style etc

What it solves

GNOS transforms a standard coding agent into a structured teaching system. It addresses the problem of AI learning experiences that feel generic, lack a coherent route, or mistake a correct answer for actual understanding. It provides the structure needed to turn a raw LLM into a personalized tutor that can diagnose mistakes and make abstract ideas visible through visual tools.

How it works

GNOS acts as a teaching harness composed of skills, scripts, and visual tools. When a user specifies a topic, the system uses course materials (like university syllabi and textbooks) to plan a learning route based on the user's desired depth and time constraints.

Subagents create individual lesson blocks, which GNOS reviews and joins together. The system records the user's attempts, reasoning gaps, and independent capabilities to dynamically adjust upcoming lessons and exercises. To move beyond text, it utilizes visual tools like diagrams, simulations, narrated animations, and PDF handouts.

Who it’s for

Developers and learners using coding agents (such as Codex, Claude Code, and Cursor) who want a structured, personalized educational course built around their current knowledge level and professional goals.

Highlights

  • Adaptive Learning: Tracks learner progress and distinguishes between solving with help versus solving independently to adjust the course.
  • Course Viewer: A dedicated browser-based tool to render the curriculum and lessons as a study page.
  • Multi-modal Explanations: Supports simulations, narrated videos, and diagrams via MCP servers (e.g., Excalidraw, Pinepaper).
  • Agent Integration: Pluggable into various AI coding agents including Codex, Claude Code, and Cursor.
  • Personalized Personas: Uses SOUL.md files to provide subject-specific teaching styles and judgment.

Related

  • Project
    jacke-pp/open-yoki

    open-yoki is a self-contained AI livestream host that runs on one server, with a cloned voice, digital avatar, chat interaction, memory, and autonomous posting. It integrates with Binance Square and TwinOasis, and is fully customizable.

  • Project
    beevibe-ai/beevibe

    An agent-native operating system that creates a shared workspace for humans and AI agents to collaborate, preventing the loss of shared intelligence across a team.

  • Project
    openma-ai/open-managed-agents

    Open Managed Agents (OpenMA) is an open‑source, self‑hostable platform for running AI agents with any model, harness, sandbox, and cloud provider. It provides durable sessions, tool integrations, and drop‑in compatible APIs for Claude Managed Agents and OpenAI Agents, plus a web console and CLI for configuration.

  • Project
    remorses/usecomputer

    usecomputer is a cross‑platform CLI (and TypeScript library) that lets AI agents take screenshots, map coordinates, and perform mouse/keyboard actions (click, drag, type, scroll) on macOS, Linux and Windows. It uses a native Zig backend, provides a coord‑map system for accurate screen‑to‑image mapping, streams global input events via SSE, and supports Kitty Graphics Protocol for agent‑friendly image delivery.