CLI Installation

Installation

bun

bun add -g @llmist/cli

npm

npm install -g @llmist/cli

pnpm

pnpm add -g @llmist/cli

yarn

yarn global add @llmist/cli

Or run directly without installing:

bunx

bunx llmist --help

npx

npx llmist --help

Verify the installation:

llmist --version

Configuration

1. Set up API keys

   Set at least one provider API key. See Environment Variables for all options.

   export ANTHROPIC_API_KEY=sk-ant-...
   # or OPENAI_API_KEY, GEMINI_API_KEY

2. Initialize configuration (optional)

   Create a configuration file for custom commands:

   llmist init

   This creates ~/.llmist/cli.toml with default settings.

3. Verify setup

   Test with a simple completion:

   llmist complete "Hello, world!"

Multiple Providers

You can set multiple API keys. The CLI will use the appropriate provider based on the model you request.

Configuration File

The configuration file at ~/.llmist/cli.toml supports:

• Default model and options for each command
• Custom commands with pre-defined prompts
• Prompt templates for reusable content
• Gadget approval rules for security

Example:

~/.llmist/cli.toml

[global]
log-level = "warn"

[complete]
model = "sonnet"
temperature = 0.7

[agent]
model = "gpt-5"
max-iterations = 10
tui = true

# Custom command
[code-review]
inherit = "agent"
system = "You are a code reviewer. Analyze the provided code for bugs, security issues, and best practices."

Next Steps

Now that you have the CLI installed, try the Quick Start to run your first agent!