> ## Documentation Index
> Fetch the complete documentation index at: https://www.runmirrors.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Mirrors documentation

> What a mirror is and where to start.

Mirrors builds a runnable digital twin of the backends an AI agent talks
to, from the agent's own traces. Your agent calls the twin instead of
production; recorded conversations replay against it to measure how
faithfully it answers; the numbers say which tool to fix and where.

## Start here

* [Quickstart](/docs/quickstart): traces in, a twin out, your agent calling it.
* [How it works](/docs/how-it-works): the words the product uses and what each measures.
* [Trace exports](/docs/traces): what a good export contains, checked before you upload.

## Ways in

* [Collectors](/docs/collectors): stream traces from a running agent (TypeScript, Python, Go).
* [The MCP server](/docs/mcp): drive everything from an IDE agent.
* [The CLI](/docs/cli): the `mirrors` command.
* [API keys](/docs/api-keys) and [plans and usage](/docs/plans-and-usage).
* [For agents](/docs/for-agents): the pages an IDE agent reads on its own.

## Reference

* [Introduction](/docs/reference/introduction), [collect](/docs/reference/collect), [the tool plane](/docs/reference/envs), [tools and calls](/docs/reference/tools), [build](/docs/reference/build).
* Generated from the product itself: [service verbs](/docs/reference/verbs), [accepted trace formats](/docs/reference/formats), [file inputs](/docs/reference/inputs), [config and overrides](/docs/reference/overrides), [getting started for agents](/docs/reference/getting-started).

A plain-text index for language-model crawlers is at [llms.txt](/docs/llms.txt).