# AgentTest

## Docs

- [Assign Variant](https://agenttest.mintlify.app/api-reference/assign-endpoint.md): Assigns a variant to a user for a specific experiment based on the configured traffic distribution. This endpoint helps in dynamically routing users to different agent variants.
- [Log Result](https://agenttest.mintlify.app/api-reference/log-endpoint.md): Logs the execution details of a specific variant run for an experiment. This data is crucial for analyzing experiment performance.
- [Assignment & Logging](https://agenttest.mintlify.app/concepts/assignment-logging.md): Two fundamental processes in AgentTest are Variant Assignment (determining which version a user sees) and Result Logging (recording what happened).
- [Authentication](https://agenttest.mintlify.app/concepts/authentication.md): The AgentTest backend uses API keys to authenticate requests. All API endpoints are protected and require a valid API key to be included in your requests.
- [Experiments & Variants](https://agenttest.mintlify.app/concepts/experiments-variants.md): AgentTest revolves around the concepts of Experiments and their Variants. Understanding these is key to effectively using the platform.
- [SDK Integration](https://agenttest.mintlify.app/developer-guide/sdk-integration.md): While the AgentTest backend exposes a clear and simple HTTP API for assigning variants and logging results, the most convenient way to integrate AgentTest into your AI agent applications will typically be through dedicated Software Development Kits (SDKs).
- [Installation & Setup](https://agenttest.mintlify.app/getting-started/installation-setup.md): Start using AgentTest with our JavaScript SDK or REST API in minutes.
- [Overview](https://agenttest.mintlify.app/index.md): Learn about AgentTest, an open-source A/B testing and analytics platform for AI agents.
- [Roadmap](https://agenttest.mintlify.app/roadmap.md): Future plans and upcoming features for AgentTest.

## OpenAPI Specs

- [openapi](https://agenttest.mintlify.app/api-reference/openapi.json)