Frontier AILago has been able to follow the pace of our releases and has allowed us to focus on what we do best.
Lago for AI agents
Build billing on Lago with your coding agent
Run the isolated local demo to see Lago turn usage into a charge. For production, Lago Cloud runs the billing service while your agent connects your application through the API or SDK. Self-hosting is optional.
Connect the tools your workflow needs
Use the API in your application, the CLI in your terminal, or MCP in a connected assistant. Choose the path that fits your workflow.
Lago CLI
Inspect customers, configure billing, and check results from your terminal. JSON output fits scripts and coding agents.
Use the CLI
Product MCP
Give your assistant connected tools for live billing data and supported operations. Configure the Lago environment and API key before use.
Connect the MCP server
API and SDKs
Build billing into your application with the REST API or an official SDK. Give your agent the OpenAPI specification for exact request and response fields.
Explore the API
Teams build their billing with AI.On Lago, from prompt to production.
See how ambitious teams build and scale their billing with Lago.
Frontier AIBefore you start
Do I need a plugin to use Lago with an agent?
No. Give your agent the public documentation, then use the REST API, CLI, or product MCP for the work you need. The optional Lago plugin is public and packages billing guidance and examples for Claude Code and Codex. See github.com/getlago/lago-plugin for installation instructions.
Do I need a Lago account to start?
You can read the public documentation and design an integration without a Lago account or API key. Checking real billing behavior requires a configured environment. Use a separate Lago Cloud organization or an isolated self-hosted instance for tests.
Do I need to self-host Lago in production?
No. Lago Cloud is a managed billing service. Your team integrates application data, configures pricing, and connects a payment provider; Lago operates the Cloud billing infrastructure. Enterprises and regulated teams can instead self-host in their own VPC, on-premises, or air-gapped environment when required.
Should I use the CLI or MCP?
Use the CLI for terminal workflows, scripts, and coding agents that can run shell commands. Use the product MCP server for assistants that call connected tools. Both access Lago through configured credentials. You do not need both.
How does my application connect to Lago?
Your application uses a Lago SDK or the REST API. The CLI and MCP give agents and operators convenient ways to work with Lago; application code does not need to run them.