Now available on npm

Prompt CLI

Fetch prompts from Promptodex, render templates with variables, and execute them against your configured AI models—all from your terminal.

$ npm install -g pod-cli

or use with npx: npx pod-cli summarize

How it works

Three simple steps to run prompts from your terminal

Install & configure

Run npm i -g pod-cli and add your API keys to ~/.pod/config.json

Run a prompt

Use pod <slug> to fetch and execute any public prompt

Get results

Output is printed to stdout—pipe it, save it, or use it in your workflow

Terminal

$ pod summarize --topic 'quantum computing'

Output:

Quantum computing leverages quantum mechanical phenomena like superposition and entanglement to process information. Unlike classical bits that are either 0 or 1, quantum bits (qubits) can exist in multiple states simultaneously, enabling exponentially faster computation for certain problems...

Features

Everything you need to run prompts from the command line

Execute any prompt

Fetch prompts from Promptodex and run them against your configured AI models instantly

Template variables

Pass variables like --topic dogs or --language spanish to fill in prompt templates

Stdin support

Pipe content directly: cat article.md | pod summarize

Multi-model support

Configure OpenAI, Anthropic, and more. Override with --model

Local caching

Prompts are cached locally at ~/.pod/cache for faster subsequent runs

Open source

MIT licensed. View the source, contribute, or fork it

Quick setup

Create a config file at ~/.pod/config.json

~/.pod/config.json

{
  "defaultModel": "4.1",
  "vendors": {
    "openai": {
      "apiKey": "sk-your-openai-key"
    },
    "anthropic": {
      "apiKey": "sk-your-anthropic-key"
    }
  },
  "models": {
    "4.1": {
      "vendor": "openai",
      "model": "gpt-4.1"
    },
    "sonnet": {
      "vendor": "anthropic",
      "model": "claude-sonnet-4"
    }
  }
}

Usage examples

Common workflows with the pod CLI

Run a prompt

Fetch and execute a prompt by its slug

$ pod summarize

Pass variables

Fill in template variables with flags

$ pod translate --language spanish --text 'Hello world'

Pipe stdin

Pipe content into the prompt's {{content}} variable

$ cat article.md | pod summarize

Override model

Use a different model than the prompt's default

$ pod code-review --model sonnet

Check configuration

Display your current settings (with masked API keys)

$ pod config

Run diagnostics

Verify your config, API keys, and registry connection

$ pod doctor

Commands

pod <slug>

Fetch and execute a prompt from Promptodex

Options

--model <alias>Override the model to use

--<variable> <value>Set template variables

-v, --verboseShow verbose output

pod config

Display configuration information including config file location and current settings

pod doctor

Run diagnostic checks: config validity, API keys, registry connection, cache access

Ready to get started?

Install the pod CLI and start running prompts from your terminal.

$ npm install -g pod-cli