Skip to content

Overview

Anote AI is a unified AI coding assistant and private chatbot platform. It reads your codebase, edits files, runs commands, reviews PRs, and answers questions over your documents — available in your terminal, IDE, browser, desktop app, and phone.

Get started

Anote runs on several surfaces: the CLI, VS Code, the web, desktop, and mobile. Pick one below to get going. Most surfaces talk to the hosted Anote backend or your own self-hosted instance (see Configuration).

The full-featured CLI for working with Anote directly in your terminal. Ask questions, fix bugs, review PRs, and search your codebase without leaving the shell.

bash npm install -g @anote-ai/anote

Requires Node.js 18 or later. Then, in any project:

bash cd your-project anote init anote ask "explain this codebase"

anote init walks you through setting your API key and preferred LLM provider.

Continue with the Quick Start →

The VS Code extension brings a chat sidebar, inline diff review, and streaming responses directly into your editor.

Search for "Anote" in the VS Code Extensions marketplace, or install via:

bash code --install-extension anote-ai.anote-ai-coding

VS Code Extension overview →

A ChatGPT-style browser chat interface with document upload and RAG-backed Q&A. Self-host it with Docker Compose:

```bash git clone https://github.com/anote-ai/Panacea cd Panacea cp packages/backend/.env.example packages/backend/.env

Edit .env with your API keys

docker compose up ```

Frontend: http://localhost:3000 · Backend: http://localhost:5000

Web App overview →

A private, offline-capable Electron app. All data stays on your machine, and it works with local Ollama models when you don't want to call out to a hosted provider.

Download the latest release from GitHub Releases — available for macOS (DMG), Windows (installer), and Linux (AppImage/DEB/RPM).

Desktop App overview →

A native iOS and Android chat client built with Expo.

bash cd packages/mobile npm install npx expo start

Scan the QR code with the Expo Go app, or run in a simulator.

Mobile App overview →

What you can do

Ask questions about your codebase

bash anote ask "how does the authentication middleware work?" anote ask --file src/auth.ts "explain this file" anote ask --compare # side-by-side across multiple models cat src/handler.py | anote ask "what could go wrong here?"

Fix bugs automatically

anote fix --loop iterates against your test suite — up to --max-iterations rounds — until it passes, or fixes a single file with --file.

bash anote fix --loop --max-iterations 5

Review pull requests

bash anote review --pr 42

Reviews for bugs, security issues, and quality — locally against a directory/file, or posted straight to a GitHub PR.

Search your codebase semantically

bash anote index # build a TF-IDF index (run once, then keep updated) anote search "JWT token validation"

Chat and Q&A over your documents

Upload documents in the Web App or Desktop App and ask questions against them — RAG-backed via POST /api/documents/{id}/ask.

Audit for security and performance issues

bash anote security --severity high --fix anote perf --focus "database,bundle" --fix

Generate changelogs and docs, or run migrations

bash anote changelog --since v1.2.0 anote docs src/api.ts --style jsdoc anote migrate --from "React 17" --to "React 18"

Check your setup

bash anote doctor

Checks Node.js ≥ 18, ANTHROPIC_API_KEY, .anote.json, CLAW.md, and git.

Use Anote everywhere

I want to... Best option
Work from my terminal CLI
Get inline AI help in my editor VS Code Extension
Chat with documents in a browser Web App
Keep everything private and offline Desktop App — works with local Ollama models
Chat from my phone Mobile App
Call Anote from my own code or scripts TypeScript SDK or Python SDK
Integrate directly against the REST API Backend API
Automate PR review or CI checks CLI: anote review --pr

Supported LLM providers

  • Anthropic — Claude (claude-opus-4-8, claude-sonnet-4-6, claude-haiku-4-5)
  • OpenAI — GPT-4o, GPT-4o-mini
  • Google — Gemini 2.0 Flash, Gemini 1.5 Pro
  • Ollama — any local model (Llama 3, Mistral, etc.)
  • xAI — Grok

Next steps