> ## Documentation Index
> Fetch the complete documentation index at: https://docs.star.fun/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

<video src="https://mintcdn.com/star-a73a9f07/KEGQ_g3yY3FV4IZP/assets/bg_video.mp4?fit=max&auto=format&n=KEGQ_g3yY3FV4IZP&q=85&s=96223155d2e4335abb180b41ccaa4a5b" autoPlay loop muted className="page-video-300" data-path="assets/bg_video.mp4" />

The CLI is the primary interface to Star. Use it to design a token, open a round, and operate rewards campaigns. The dashboard shows the same data, but the CLI controls the platform.

## Install

Requires Node.js 22 or 24. Install the exact current release, `0.4.0`:

```bash theme={null}
npm install --global @star-factory/launch-cli@0.4.0
star --version   # prints 0.4.0
```

Or run it without installing:

```bash theme={null}
npx -y @star-factory/launch-cli@0.4.0 --help
```

Pin the version. An unpinned `@star-factory/launch-cli` resolves the npm `latest` tag, which can lag the current release and hand you an older command surface.

## Launch from the terminal

Every launch carries four founder decisions, and the CLI defaults none of them: the raise tier, the equity link percent, whether to create the compensation-band plan, and whether to reserve the Launch Rewards allocation. Start by reading what the deployment requires, then authorize, preview, and launch.

```bash theme={null}
# 1. Read the requirements: the four decisions, Star's prompt for each, and what every tier raises and graduates at
star launch requirements

# 2. Authorize this device (prints a star.fun URL + user code)
star launch login

# 3. Preview exactly what will be sent (no mutation)
star launch create \
  --name "My Startup" --ticker MYSTAR --website https://mystartup.com --icon ./logo.png \
  --tier 25k --equity-link 10 --compensation-bands --launch-rewards \
  --dry-run --json

# 4. Launch for real and wait until the token is live
star launch create \
  --name "My Startup" --ticker MYSTAR --website https://mystartup.com --icon ./logo.png \
  --tier 25k --equity-link 10 --compensation-bands --launch-rewards \
  --yes --wait-until live --timeout 30m --json
```

The tier ids, equity-link bounds, and each tier's economics come from `star launch requirements`, never from the CLI itself. Leave a decision out and `create` asks for it on a terminal; with `--json` it returns `DECISION_REQUIRED` naming the unanswered questions. `--compensation-bands` and `--launch-rewards` each have a `--no-` form.

Production is the default. To target staging, pass `--env staging` on every command.

## Use it from Codex or Claude

The package is three pieces: the `star` CLI, the `launch-token-on-star` skill (the interview that asks the four founder decisions in Star's words), and the local MCP server (the tools). `mcp add` alone registers tools. It does not install the skill. The native plugin that would do both is not listed in the Codex or Claude marketplaces yet, so install all three:

**Claude Code**

```bash theme={null}
npm install --global @star-factory/launch-cli@0.4.0
mkdir -p ~/.claude/skills
cp -R "$(npm root -g)/@star-factory/launch-cli/skills/." ~/.claude/skills/
claude mcp add --scope user star -- npx -y @star-factory/launch-cli@0.4.0 mcp serve
claude mcp list
```

`--scope user` makes the MCP server available in every project. Claude's default is the current project only.

**Codex CLI**

```bash theme={null}
npm install --global @star-factory/launch-cli@0.4.0
mkdir -p ~/.agents/skills
cp -R "$(npm root -g)/@star-factory/launch-cli/skills/." ~/.agents/skills/
codex mcp add star -- npx -y @star-factory/launch-cli@0.4.0 mcp serve
codex mcp list
```

Codex writes the MCP entry to `~/.codex/config.toml`, so it is available in every project.

If the CLI is already installed at `0.4.0`, skip the `npm install` line and keep the skill copy plus `mcp add`.

**Cursor, or any client with an `mcp.json`**

```json theme={null}
{
  "mcpServers": {
    "star": {
      "command": "npx",
      "args": ["-y", "@star-factory/launch-cli@0.4.0", "mcp", "serve"]
    }
  }
}
```

Then copy the skills from the installed package into that host's skill directory the same way. Every MCP tool accepts a per-call `environment` (`production` or `staging`); without one it targets production.

## Launch with your agent

1. Ask the agent to launch. Example: "Launch my token on Star. Name My Startup, ticker MYSTAR, website mystartup.com."
2. The agent reads the launch requirements and asks you the four founder decisions in Star's own words. It never picks them for you.
3. The agent starts a secure sign-in. Open the link, sign in with your Star account, and enter the code.
4. Star derives your launch wallet and reserves a Star vanity mint.
5. Examine the exact payload and approve it. Nothing launches before you approve.

Agents do not have access to wallet secrets or keypairs.

## Capabilities

The CLI launches tokens, opens [curve rounds](/founders/fundraising), and operates [rewards campaigns](/founders/rewards) today. The rest is on its way:

| Capability                                            | Status                                    |
| ----------------------------------------------------- | ----------------------------------------- |
| Presales and pre-commit campaigns                     | On request. CLI support is in development |
| Token utility                                         | In development                            |
| [Revenue share](/founders/revenue-share) and buybacks | On request. CLI support is in development |

**On request**: we configure the feature with you. [Book a call](https://cal.com/adam-bergeman/founder-call).

**In development**: the documentation shows the planned behavior. You can plan your integration now.

## Example session

```text theme={null}
> design a token with 20% revshare and in-app credit utility
> open a round on Star
> reward the top 100 referrers weekly from the rewards pool
```

The CLI validates your configuration. It shows the full plan and waits for your approval before launch. Read [Curves](/founders/fundraising) for the raise mechanics.

## Give these docs to your agent

The docs are agent-readable. Point your agent at [docs.star.fun/llms.txt](https://docs.star.fun/llms.txt) for the index, or [llms-full.txt](https://docs.star.fun/llms-full.txt) for the full text. Append `.md` to any page URL to get that page as plain markdown. The full command surface is in the [CLI reference](/reference/cli).