akan-team/akanjs

The TypeScript framework, agents included. Web, iOS, Android, macOS, Windows and Linux. Your server and database. And the AI agents your users already talk to.

What it solves

Akan.js is a full-stack TypeScript framework designed to bridge the gap between human-facing web/mobile applications and AI agents. It eliminates the need to manually write separate tool schemas or MCP (Model Context Protocol) servers by automatically exposing your application's business logic, UI controls, and API endpoints as AI-ready tools.

How it works

The framework uses a unified "one-line" declaration system where a single definition of a data field or business action propagates across eight layers, including schemas, API endpoints, and UI components. Because your app is built with these shared definitions, the framework automatically publishes your server endpoints as MCP tools and your screen controls as in-page agent tools. It enforces strict security guards that apply equally to human users and AI agents, ensuring that an AI can only perform actions that the current user is authorized to execute.

Who it’s for

Developers building cross-platform applications (web, iOS, Android, and desktop) who want their software to be natively "agent-aware." It is particularly useful for teams who want to integrate AI capabilities into their apps without maintaining separate, complex tool definitions or permission models for LLMs.

Highlights

  • Unified Agent Interface: Every UI control and guarded server endpoint is automatically exposed as an AI tool, allowing agents to interact with the app as a user would.
  • Cross-Platform Support: Ships the same codebase to web, mobile, and desktop environments with native-level performance.
  • Security-First Design: Uses the same authorization guards for humans and AI, ensuring agents cannot access unauthorized data or perform restricted actions.
  • Agent-Optimized Workflow: Includes built-in support for coding agents (like Claude Code or Cursor) with pre-configured AGENTS.md rules and MCP integration to keep code consistent.
  • High Performance: Built on Bun, offering fast startup times and efficient memory usage.

Related

  • Project
  • Project
  • Project
  • Project