stripe-token-billing

CLI tool for managing Stripe Rate Card models for token-based billing.

Automates the process of adding new LLM models to a Stripe Rate Card by creating Metered Items and Rates for both input and output tokens in a single command.

Setup

1. Install

npx stripe-token-billing --help

2. Configure credentials

Create config files for each environment under ~/.config/stripe-token-billing/:

mkdir -p ~/.config/stripe-token-billing

Sandbox (~/.config/stripe-token-billing/.env.sandbox):

STRIPE_SECRET_KEY=sk_test_...
STRIPE_RATE_CARD_ID=brc_...
STRIPE_METER_ID=mtr_...

Production (~/.config/stripe-token-billing/.env.production):

STRIPE_SECRET_KEY=sk_live_...
STRIPE_RATE_CARD_ID=brc_...
STRIPE_METER_ID=mtr_...

Environment Variables

Variable	Description
STRIPE_SECRET_KEY	Stripe API key (sk_test_... or sk_live_...)
STRIPE_RATE_CARD_ID	Existing Rate Card ID (rcd_...)
STRIPE_METER_ID	Existing Meter ID (mtr_...)

Usage

The --env flag is required for all commands. It loads the corresponding config file (~/.config/stripe-token-billing/.env.<environment>).

Add a model

# Preview what will be created (no API calls)
npx stripe-token-billing --env sandbox add \
  --model "google/gemini-3-flash" \
  --input-price 0.50 \
  --output-price 3.00 \
  --markup 10 \
  --dry-run

# Create in sandbox
npx stripe-token-billing --env sandbox add \
  --model "google/gemini-3-flash" \
  --input-price 0.50 \
  --output-price 3.00 \
  --markup 10

# Then create the same model in production
npx stripe-token-billing --env production add \
  --model "google/gemini-3-flash" \
  --input-price 0.50 \
  --output-price 3.00 \
  --markup 10

List rates

npx stripe-token-billing --env sandbox list

Options

Option	Description	Default
--env	Environment to use (e.g. sandbox, production)	Required
--model	Model identifier (e.g. google/gemini-3-flash)	Required
--input-price	Price per million input tokens (USD)	Required
--output-price	Price per million output tokens (USD)	Required
--markup	Markup percentage	10
--dry-run	Preview without making API calls	false

Development

pnpm install
pnpm test          # Run tests
pnpm test:watch    # Run tests in watch mode
pnpm format        # Auto-fix lint & format
pnpm build         # Compile TypeScript

What it creates

For each model, the add command creates:

1. Input tokens Metered Item — with meter_segment_conditions for model and token_type=input
2. Input tokens Rate — unit amount = input_price × (1 + markup%) ÷ 1,000,000
3. Output tokens Metered Item — same as above with token_type=output
4. Output tokens Rate — unit amount = output_price × (1 + markup%) ÷ 1,000,000

All resources are tagged with created_by: "Stripe Token Billing" metadata.