This guide is for people who want the easiest setup path.
You do not need to build from source. You do not need Bun. You do not need to understand the full codebase.
If you can copy and paste commands into a terminal, you can set this up.
OpenClaude lets you use an AI coding assistant with different model providers such as:
- OpenAI
- DeepSeek
- Gemini
- Ollama
- Codex
For most first-time users, OpenAI is the easiest option.
You need:
- Node.js 22 LTS or newer installed
- A terminal window
- An API key from your provider, unless you are using a local model like Ollama
- Install OpenClaude with npm
- Run
openclaude - Inside the CLI, run
/providerfor guided provider setup
The /provider command walks you through choosing a provider and entering credentials. You do not need to set environment variables beforehand.
- Windows: Windows Quick Start
- macOS / Linux: macOS / Linux Quick Start
Once you have picked a provider, run /provider inside OpenClaude to set it up with guided prompts.
Choose this if:
- you want the easiest cloud setup
- you already have an OpenAI API key
Choose this if:
- you want to run models locally
- you do not want to depend on a cloud API for testing
Choose this if:
- you already use the Codex CLI
- you already have Codex or ChatGPT auth configured
After you run openclaude, the CLI should start and wait for your prompt.
At that point, you can ask it to:
- explain code
- edit files
- run commands
- review changes
Cause:
- npm installed the package, but your terminal has not refreshed yet
- on Windows, npm's global bin folder may not be in your user
Path
Fix:
- Close the terminal
- Open a new terminal
- Run
openclaudeagain
On Windows PowerShell, if that still does not work, add npm's global bin folder
to your user Path, then open a new PowerShell window:
$npmPrefix = npm config get prefix
$currentUserPath = [Environment]::GetEnvironmentVariable("Path", "User")
if (($currentUserPath -split ';') -notcontains $npmPrefix) {
[Environment]::SetEnvironmentVariable(
"Path",
"$currentUserPath;$npmPrefix",
"User"
)
}Cause:
- the key is wrong, expired, or copied incorrectly
Fix:
- Get a fresh key from your provider
- Run
/providerinside OpenClaude to update your credentials - Re-run
openclaude
Cause:
- OpenClaude does not automatically load
.envfiles. If you copied.env.exampleto.env, OpenClaude won't see the variables unless you tell it to.
Fix:
- Load the file explicitly:
openclaude --provider-env-file .env - Or, use the
/providercommand inside OpenClaude instead (recommended). - Do not commit your
.envfile to git. - The explicit loader accepts provider/setup variables. Export runtime/debug variables from your shell or launcher instead.
Cause:
- Ollama is not installed or not running
Fix:
- Install Ollama from
https://ollama.com/download - Start Ollama
- Try again
If you want source builds, advanced provider profiles, diagnostics, or Bun-based workflows, use:
- Advanced Setup This is also where to find Codex, Gemini, Mistral, LiteLLM, and profile-launcher setup.
-
GitHub Discussions: https://github.com/Gitlawb/openclaude/discussions Use this for Q&A, setup help, and community conversation.
-
GitHub Issues: https://github.com/Gitlawb/openclaude/issues Use this for confirmed bugs and feature requests.
If OpenClaude is not working after setup, run:
openclaude --versionIf this prints a version number, the install succeeded. If it says "command not found," close your terminal, open a new one, and try again. On Windows, you may also need to add npm's global bin folder to your user Path (see the Windows Quick Start guide for details).
When filing a bug, run this and paste the redacted output into the issue:
openclaude doctor report --markdown