# Evaris Docs

> Evaris is an evaluation platform for AI agents: run Inspect AI evals, publish the logs, and review runs in a web UI. Every page below is also served as raw markdown by appending `.md` to its URL.

## Getting Started

- [Authentication](https://evarisai.com/docs/getting-started/authentication): Evaris has two authentication paths. They are used at different stages of the
- [Install](https://evarisai.com/docs/getting-started/install): Publishing eval logs to Evaris needs two toolchains: Python for running Inspect
- [Overview](https://evarisai.com/docs/getting-started/overview): Evaris is an evaluation platform for AI agents. You run an evaluation, today
- [Quickstart](https://evarisai.com/docs/getting-started/quickstart): End-to-end: sign up, mint a token, run a sample Inspect eval, publish it, see
## Guides

- [Inspect AI integration guide](https://evarisai.com/docs/guides/inspect-ai-integration): How to author an eval with [Inspect AI](https://inspect.aisi.org.uk/), run it,
- [OpenAI-compatible providers](https://evarisai.com/docs/guides/openai-compatible-providers): Inspect AI talks to any OpenAI-compatible endpoint. This page covers the model
- [Publish eval logs](https://evarisai.com/docs/guides/publish-eval-logs): Deep dive on how publishing works: the publishInspectLogs function, the
## Reference

- [REST API reference](https://evarisai.com/docs/reference/api): The Evaris backend exposes a small REST API; this page is a human-readable
- [CLI reference](https://evarisai.com/docs/reference/cli): The evaris CLI publishes Inspect AI eval logs to a Evaris backend. It is the
- [Data model](https://evarisai.com/docs/reference/data-model): The shapes the SDK and API exchange. These mirror the package's generated
- [TypeScript SDK reference](https://evarisai.com/docs/reference/sdk-typescript): The evaris package is a TypeScript client for the Evaris backend. It has
## Examples

- [Example: custom solver (no LangChain)](https://evarisai.com/docs/examples/custom-solver): The LangChain example ([previous page](./support-agent-eval.md)) is one
- [Example: LangChain support/refund agent eval](https://evarisai.com/docs/examples/support-agent-eval): A walkthrough of the reference example. It evaluates a LangChain

## Optional

- [Full docs dump](https://evarisai.com/llms-full.txt): every page in one file