Skip to main content
Weflayr is set up in code, so a coding agent (Claude Code, Cursor, Codex, Copilot, …) can do the integration for you: install the SDK, auto-instrument your AI providers, wrap your calls with the right feature and customer, and hand you back what is left to configure.

Paste this to your agent

What the agent will do

1

Survey your code

Finds your AI provider SDKs, your framework, every process that makes LLM calls, and where your customer or tenant identity lives.
2

Install and auto-instrument

Adds the SDK and one startup call per process, placed so the provider SDKs are patched before they are used.
3

Name your features and wrap your calls

Proposes a feature name per product-side job, then wraps each one with a decorator or a context manager, whichever fits your code.
4

Fill in your customers

Takes the customer identifier from your code when it is there, and asks you when it is not.
5

Ask you the rest

Where your revenue comes from, whether you have free AI credits, whether this is a product or an internal tool, and what should happen with staging.
6

Summarize its changes & explain next steps

What it changed, what you still need to do in the dashboard, and what it could not do.

Also machine-readable

Documentation index

Every page of this site, as a list an agent can fetch.

Full documentation

Every page of this site, as one file.

Weflayr MCP

Your cost, margin and revenue data, queried by your assistant.

Any page as markdown

Add .md to any documentation URL to get its source.