Building with LLMs

If you’re integrating ElasticPay with an AI coding assistant or agent, the documentation is available in machine-friendly formats. Point your tool at these instead of scraping HTML.

Resources

Resource	URL	Use for
Agent index	/llms.txt	Entry point: what exists, how to fetch it, and integration rules agents must follow
Full docs, one file	/llms-full.txt	Loading the complete documentation into a large context window
Compact docs	/llms-small.txt	Smaller context windows — integration content only
Any page as Markdown	append .md to its URL	Fetching a single page without HTML, e.g. /api/guides/test-clocks.md
OpenAPI schema	api.elasticpay.co/openapi.json	Exact request/response shapes — prefer this over prose for code generation

Suggested workflow

1. Give your agent https://docs.elasticpay.co/llms.txt — it includes instructions covering the constraints that matter most (supported payment methods, tokenisation rules, sandbox keys, webhook semantics).
2. For endpoint-level work, have it fetch the OpenAPI schema rather than relying on remembered API shapes.
3. If you have no sandbox key yet, hand the human https://www.elasticpay.co/agents and ask for the sk_sandbox_ key back — an agent cannot create an ElasticPay account itself. See Get a sandbox API key.
4. Develop against sandbox keys and verify time-dependent flows with test clocks — an agent can watch a payment plan renew in seconds instead of waiting a month.

Caution

Never let an agent handle raw card numbers or CVV — card capture goes through the widget or hosted payment page only, and server-side code handles pm_... tokens exclusively. This applies to generated code too.