Portal Guide & AI Tools | Algorand Developer Portal

Portal Guide & AI Tools

Welcome to the Algorand Developer Portal. Whether you have been waiting for this or are surprised by it, please have a look around. We think you’re going to enjoy reading — or chatting with — these docs.

What You’ll Find Here

Build with AI

The Algorand developer ecosystem provides AI tools that help you code faster, find answers instantly, and follow best practices. Whether you prefer chatting in the browser or using an AI assistant in your IDE (Integrated Development Environment), there are options for you.

Ask AI on This Site

Click the Ask AI button on any page to chat with an AI assistant that has been trained on the full contents of this portal. It can answer questions about Algorand concepts, code patterns, and tooling.

Kapa MCP Server

The same AI that powers the chatbot on this site is available as an MCP server, bringing Algorand documentation search directly into your coding assistant.

Add the Kapa MCP server to your configuration:

Add to .mcp.json in your project root:

{
  "mcpServers": {
    "algorand-docs-kapa": {
      "type": "http",
      "url": "https://algorand-docs.mcp.kapa.ai/"
    }
  }
}

Add to .cursor/mcp.json in your project:

{
  "mcpServers": {
    "algorand-docs-kapa": {
      "url": "https://algorand-docs.mcp.kapa.ai/"
    }
  }
}

Add to ~/.codex/config.toml or .codex/config.toml in your project:

[mcp_servers.algorand-docs-kapa]
url = "https://algorand-docs.mcp.kapa.ai/"

Add to .vscode/mcp.json in your project:

{
  "servers": {
    "algorand-docs-kapa": {
      "type": "http",
      "url": "https://algorand-docs.mcp.kapa.ai/"
    }
  }
}

Add to opencode.json in your project root:

{
  "mcp": {
    "algorand-docs-kapa": {
      "type": "remote",
      "url": "https://algorand-docs.mcp.kapa.ai/",
      "enabled": true
    }
  }
}

Algorand Agent Skills

Algorand Agent Skills are a collection of skill files that teach AI coding assistants how to build on Algorand. They cover the full development lifecycle — from project initialization and smart contract development to testing, deployment, and troubleshooting. Skills are available for Claude Code, Cursor, OpenCode, and GitHub Copilot.

LLM-Optimized Content

If you want to feed Algorand documentation directly into your AI tools, the full site content is available in plain text:

About This Portal

In reimagining the Algorand Developer Portal, we set out to address challenges from the past, cater to what developers expect today, and plan for how technical documentation will be used in the future. Our work has been guided by a set of key principles:

  1. One-stop shop: The portal should consolidate all of the resources developers need to build with Algorand technologies. If something cannot be found, it may as well not exist. Therefore, the portal should colocate documentation and examples for all of our tools, libraries, and languages under a single site.
  2. Correct and current: The content of the documentation needs to accurately reflect the current state of the art of Algorand code and tools.
  3. Diataxis framework: These docs have been designed loosely following the Diataxis approach to technical documentation, which conceptualizes content along the axes of action vs. cognition and acquisition vs. application. The four quadrants which emerge from a plot of these axes are:
    • Tutorials, for learning, such as in the Getting Get Ready To Build section
    • How-To Guides, for accomplishing specific goals, such as the Running a Node section
    • Explanation, for understanding concepts, in the Concepts section
    • Reference, for finding information, in the Reference section

This site is built using the Starlight documentation theme for Astro, an industry-leading static site generation framework. The integrated AI bot is provided by kapa.ai.

Feedback

If you’re reading this portal, then the portal is for you. Whether you are actively building applications on Algorand, performing formal research, or just following your curiosity, we appreciate your feedback. All our docs are open source under the Algorand Foundation organization on GitHub.

Submit Feedback Report issues, suggest improvements, or contribute to the documentation.