Elisa offers a novel block-based visual programming environment designed to empower users, including children, to create software by defining goals, features, and constraints. AI agents then collaborate to translate these visual specifications into functional code. This approach democratizes software design by abstracting away complex coding, providing a transparent and interactive AI-driven development experience.

How It Works

Elisa employs an agentic spec-driven development paradigm. Users define software requirements through a visual block-based interface, which generates a NuggetSpec JSON schema. A meta-planner decomposes this spec into a task DAG, orchestrating a team of specialized AI agents (builder, tester, reviewer) that execute tasks collaboratively. The entire process, including planning, coding, testing, and deployment, is made visible in real-time, with options for human-in-the-loop checkpoints and interactive agent communication.

Quick Start & Requirements

Installers are available for Windows (.exe) and macOS (.dmg) from the Releases page. Prerequisites include Node.js (LTS) and an Anthropic API key. For development from source, clone the repository, run npm install, and then npm run dev:electron to launch the Electron app. Note that the app is not code-signed, which may trigger OS security warnings. Development requires Node.js 20+ and an Anthropic API key set as an environment variable (ANTHROPIC_API_KEY). Optional dependencies include Python 3.10+ and gcloud CLI.

Highlighted Details

• Block-Based Spec: Visually design software using 14 categories of blocks, translating directly to AI engineering primitives like Goals, Requirements, Skills, Rules, Portals, and Minions.
• AI Agent Collaboration: A meta-planner orchestrates specialized AI agents (builder, tester, reviewer, custom) that communicate and collaborate to fulfill the user's spec.
• Live Build Visibility: Monitor the entire development process through a tabbed interface showing the block editor, task DAG, agent communications, test results, and system I/O.
• Hardware & IoT Integration: Directly target ESP32 boards for flashing and build multi-device IoT sensor networks communicating via LoRa and WiFi to cloud dashboards.
• Human-in-the-Loop & Debugging: Integrate user approval gates ("check with me") and utilize a "Fix It" workflow for targeted bug resolution without full rebuilds.

Maintenance & Community

No specific details regarding maintainers, community channels (e.g., Discord, Slack), or roadmap were found in the provided README.

Licensing & Compatibility

The project is licensed under the MIT License. This license generally permits commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

The application is not code-signed, necessitating manual user intervention to bypass OS security warnings on initial launch. Development and core functionality are dependent on an Anthropic API key, introducing external service costs and potential rate limits.