Kanwas provides a collaborative workspace designed for AI-assisted workflows, enabling teams and AI agents to interact seamlessly over shared documents, evidence, and decisions. It targets founders, product managers, developers, marketers, and sales teams, aiming to reduce context-switching overhead and accelerate the generation of structured, execution-ready artifacts by leveraging AI alongside human collaborators.

How It Works

Kanwas functions as a multiplayer workspace where AI agents and human teams operate on a unified "canvas." Core to its design is the concept of shared context, allowing AI tool calls and team decisions to stream into a common timeline, ensuring transparency. This approach facilitates iterative work and compounds context over time, making subsequent tasks more efficient. The system utilizes a Git-backed markdown filesystem, offering version history and mitigating vendor lock-in.

Quick Start & Requirements

• Primary Install: Docker and Docker Compose are required.
• Prerequisites: An Anthropic API key and/or OpenAI API key is necessary.
• Run Command: Clone the repository (git clone https://github.com/kanwas-ai/kanwas.git), navigate into the directory (cd kanwas), copy and fill in environment example files (.env, backend/.env, yjs-server/.env, frontend/.env), and run docker-compose --profile app up.
• Access: The application is accessible at http://localhost:5173.
• Docs: Architectural details and development setup are available in docs/SYSTEM_OVERVIEW.md.

Highlighted Details

• Multiplayer workspace for AI-driven team collaboration.
• AI agent tool calls stream directly into a shared timeline.
• Git-backed markdown filesystem ensures version history and data portability.
• Enables rapid generation of structured, execution-ready project artifacts.

Maintenance & Community

• Community: The project maintains a "Kanwas Kollective" community on Slack for questions and discussions.
• Contributing: Contributions are welcomed; contributors should review docs/SYSTEM_OVERVIEW.md, open issues for significant changes, format code with pnpm format, lint relevant packages, and sign a Contributor License Agreement (CLA) upon first contribution.

Licensing & Compatibility

• License: Kanwas is licensed under the Apache License 2.0.
• Compatibility: The Apache 2.0 license is permissive and generally compatible with commercial use and linking within closed-source projects.

Limitations & Caveats

The provided README does not explicitly detail known limitations, alpha/beta status, or specific unsupported platforms. Users should consult the docs/SYSTEM_OVERVIEW.md for deeper technical insights and potential caveats related to its core technologies like Yjs and BlockNote.