npx ainf-gpt (AINF.GPT)

Goal: let customers run npx ainf-gpt to complete setup automatically on Windows / macOS / Linux.

What it does:
- Opens the Dashboard so the user can copy the 12‑char pairing token.
- Uses that pairing token to fetch the latest active key.
- Stores the pairing token locally for reuse (device binding).
- Creates a timestamped backup of existing local settings before writing new ones.
- Optionally installs the official Codex CLI (@openai/codex).

Notes:
- npx requires Node.js LTS to be installed first.

Usage

``bash
npx ainf-gpt
npx ainf-gpt --pair-token XXXXXXXXYYYY
`

Local test (server/dev machine)

`bash
cd tools/ainf-gpt-npx
npm pack
npx ./ainf-gpt-*.tgz --pair-token XXXXXXXXYYYY --no-install
`

You can also run it directly:

`bash
node bin/ainf-gpt.js --pair-token XXXXXXXXYYYY --no-install
`

Publish to npm (so customers can run npx)

`bash
cd tools/ainf-gpt-npx
npm login
npm publish --access public
`

After publishing, customers can run:

`bash
npx ainf-gpt


Branded uppercase command (npm package names must be lowercase, so we explicitly pick the package)


npx -p ainf-gpt AINF-GPT
`

Flags

- --pair-token <12-chars>: use the 12‑char pairing token shown in Dashboard
- --origin : override API origin (useful for local/test)
- --no-open: do not auto-open the dashboard page
- --no-install: do not auto-install Codex CLI (default tries when missing)
- --install: explicitly attempt Codex CLI install (skips if already installed)
- --show-keys`: when multiple tokens exist, display full tokens in the list (default masks)