Installation

Install OpenDash via npm or the install script.

# npm (recommended)
$ npm install -g @opendash-ai/cli

# verify
$ opendash --version
0.1.0-beta.11

# macOS / Linux / WSL
$ curl -fsSL https://opendash.ai/install | bash

# windows (powershell)
PS> iwr https://opendash.ai/install.ps1 | iex

Or grab a binary directly from GitHub Releases.

Quick start

1. Install OpenDash. 2. Run opendash to start the TUI. 3. Run /connect and enter your API key. 4. Start asking questions.

$ opendash

# In the TUI, type:
/connect     # connect a provider
/model       # switch models

OpenDash requires no account. You bring your own API key from any supported provider.

Setting up providers

OpenDash supports multiple AI providers. Each provider requires an API key obtained directly from the provider.

• OpenAI — /connect → select OpenAI → enter API key
• Anthropic — /connect → select Anthropic → enter API key
• Google Gemini — /connect → select Google → enter API key
• OpenRouter — /connect → select OpenRouter → enter API key
• Local (Ollama) — Install Ollama, pull a model, then /connect → Ollama

See Providers for the full list.

Local models

Run models on your own hardware with Ollama or LM Studio. No API key required.

# Install Ollama, then:
$ ollama pull llama3
$ opendash
# In TUI: /connect → Ollama

Configuration

OpenDash uses config.json in ~/.opendash/. Key settings:

{
  "model": "gpt-4o",
  "provider": {
    "openai": { "options": { "baseURL": "..." } }
  }
}

CLI commands

MCP (Model Context Protocol)

OpenDash supports MCP servers for extending capabilities. Configure them via /mcp or in your config file.

Troubleshooting

• Models not showing after /connect — Run /model to refresh. Models are discovered from the provider's API.
• CLI not found — Ensure ~/.opendash/bin is in your PATH.
• Build errors — Run bun typecheck from the package directory.