Summary

Routa addresses the inefficiency of single-agent AI development by introducing a workspace-first, multi-agent coordination platform. It enables AI development teams to leverage specialized agents orchestrated via a Kanban board, mirroring effective human team structures for enhanced productivity and workflow visibility.

How It Works

The core mechanism employs a Kanban board as the central coordination layer. User goals are decomposed into cards, which specialized agents, bound to specific Kanban columns (Backlog, Todo, Dev, Review, Done), automatically pick up and process. This facilitates a structured flow where agents specialize, hand off work, and maintain visibility. Routa operates on dual runtime surfaces—a Next.js web application and a Tauri + Axum desktop application—sharing a unified workspace-scoped coordination model and featuring an ACP orchestration layer for normalized agent provider interaction.

Quick Start & Requirements

• Installation: Desktop (recommended) involves npm install --legacy-peer-deps, npm --prefix apps/desktop install, npm run tauri:dev. Web demo: npm install --legacy-peer-deps, npm run dev (access http://localhost:3000). Docker: docker compose up --build. CLI: npm install -g routa-cli.
• Prerequisites: Node.js/npm, TypeScript, Next.js, Rust (Axum), Tauri. Docker for containerized deployment.
• Links: Community Slack: https://join.slack.com/t/routa-group/shared_invite/zt-3txzzfxm8-tnRFwNpPvdfjAVoSD6MTJg. Demos available via Bilibili and YouTube links in README.

Highlighted Details

• Specialized Agents: Built-in agents (Refiner, Orchestrator, Crafter, Guard, Reporter, Blocked Resolver) automate distinct development stages.
• Customization: Define custom specialists via Web UI, REST API, or Markdown files.
• Provider Agnosticism: Supports multiple AI backends (Claude Code, OpenCode, Codex, Gemini CLI, Kimi, Augment, Copilot) via the Agent Client Protocol (ACP).
• Protocols: Integrates MCP, ACP, A2A, and AG-UI protocols for coordination, agent management, inter-agent communication, and UI rendering.
• GitHub Integration: Features a GitHub Virtual Workspace for browsing and code review without local cloning.
• Harness Engineering: Emphasizes AI readability, guardrails, and rapid feedback loops in system design.

Maintenance & Community

A Slack community is available for engagement. Bug reports and feature requests are managed via GitHub issues. Contribution guidelines are provided.

Licensing & Compatibility

Licensed under the MIT license, permitting broad commercial use and integration with closed-source projects.

Limitations & Caveats

The web version is explicitly designated for demonstration purposes only. Development and building require a Node.js/npm environment and the Rust toolchain.