The Sovereign
Developer Toolkit

TF Code for agentic development, MCP for AI assistant connectivity,
a sovereign API, SDKs for custom integrations, and CLI tools for automation.

TfCode - AI Coding Agent

ToothFairyAI's official terminal-based AI coding assistant. Edit files, run commands, and build features with seamless integration to your ToothFairyAI workspace.

Terminal-Based AI AssistantFile Editing & RefactoringShell Command ExecutionToothFairyAI Workspace Sync

@toothfairyai/tfcode

Install globally to start coding with AI in your terminal

$ tfcode

 _____ _____             _      
|_   _|  ___|__ ___   __| | ___ 
  | | | |_ / __/ _ \ / _` |/ _ \
  | | |  _| (_| (_) | (_| |  __/
  |_| |_|  \___\___/ \__,_|\___|

● Tip Press Tab to toggle between Plan and Task mode

Ask anything... "Fix the failing tests"

How TF Code compares with Claude Code →

Model Context Protocol (MCP)

Connect AI assistants directly to ToothFairyAI documentation and API resources. Access 47+ documentation pages, 90 API endpoints, and powerful search capabilities.

47+ Documentation Pages90 API EndpointsFull-Text SearchClaude Code IntegrationClaude Desktop Support

Local MCP (stdio)

Run the MCP server locally on your machine using stdio transport

pip install tf-mcp-server

Hosted MCP (Global)

Connect any MCP-compatible client directly to the hosted SSE endpoint — no install required

https://mcp.toothfairyai.com/sse

API Documentation

Explore our comprehensive REST API documentation. Build powerful integrations with full access to agents, knowledge hubs, conversations, and more.

Interactive API ExplorerOpenAPI SpecificationCode ExamplesAuthentication Guide

REST API Reference

Complete API reference with interactive examples, authentication guides, and endpoint documentation.

Sovereign LLM API — models & per-1M-token pricing → · full model catalog with capabilities & regional health →

Software Development Kits

Build custom applications powered by ToothFairyAI. Full programmatic access to agents, knowledge hubs, and AI capabilities for your own software.

Python SDK

Embed AI agents in Python apps, build data pipelines, and create custom integrations

pip install toothfairyai

Node.js SDK

Add AI capabilities to web apps, build APIs, and create server-side integrations

npm install @toothfairyai/sdk

Command Line Interface

Run ToothFairyAI agents directly from your terminal. Stream responses, execute automation scripts, and integrate AI into your shell workflows.

Python CLI

Run agents from terminal, stream responses, and automate tasks with shell scripts

pip install toothfairy-cli

Node.js CLI

Execute agents via npx, integrate with npm scripts, and build automation pipelines

npm install @toothfairyai/cli

Frequently Asked Questions

We've answered some of the most common questions about ToothFairyAI below. If you have any other questions, please get in touch.

What is TF Code and how do I get started?

TF Code is ToothFairyAI's sovereign AI coding agent for the terminal, TUI and web. Install it globally with npm install -g @toothfairyai/tfcode, run tfcode in any project, and connect it to your workspace with /connect. It can edit files, run commands, use MCP tools and hand off sessions to the cloud — all under your own sovereign tenancy.

What can I build with the ToothFairyAI API?

Anything the platform can do: create and run agents, query knowledge hubs, generate charts and content, manage workspaces and stream responses. The API is OpenAI-compatible, served from sovereign endpoints in Australia, Europe and the US, so existing OpenAI SDK code works with a base-URL change.

Which languages do the SDKs support?

Official SDKs are available for Python (pip install toothfairyai) and Node.js (npm install @toothfairyai/sdk). Both expose the full REST API with typed helpers, streaming support and webhook utilities. For other languages you can call the REST API directly.

What is the MCP server and which assistants does it work with?

The Model Context Protocol server gives AI assistants direct access to ToothFairyAI documentation and API resources — 47+ documentation pages, 90 API endpoints and full-text search. It works with Claude Code, Claude Desktop and any MCP-compatible client, either locally via pip install tf-mcp-server or through the hosted global endpoint.

Do the CLI tools require an account?

Yes — the CLIs authenticate against your ToothFairyAI workspace with an API key, so every call is billed to your UoI credits and respects your workspace permissions. You can generate an API key from the workspace settings in the app.

Can I self-host or run these tools in my own cloud?

Serverless customers consume the hosted sovereign endpoints. Under an Enterprise agreement, the platform — including API, agents and models — can be deployed on-premise or in your own cloud, with the same SDKs, CLI and TF Code workflows pointing at your private endpoints.

The SovereignDeveloper Toolkit