---
title: DigitalOcean Evaluations Reference
description: Native and third-party tools, troubleshooting, and answers to frequently asked questions.
product: Evaluations
url: https://docs.digitalocean.com/products/evaluations/reference/
last_updated: "2026-10-12"
---

> **For AI agents:** The documentation index is at [https://docs.digitalocean.com/llms.txt](https://docs.digitalocean.com/llms.txt). Markdown versions of pages use the same URL with `index.html.md` in place of the HTML page (for example, append `index.html.md` to the directory path instead of opening the HTML document).

# DigitalOcean Evaluations Reference

DigitalOcean Evaluations measure model, Inference Router, and agent performance using dataset evaluations and simulations.

## The DigitalOcean API

The [DigitalOcean API](https://docs.digitalocean.com/reference/index.html.md) lets you manage resources programmatically with standard HTTP requests. All actions available in the control panel are also available through the API.

The [Agent Platform API](https://docs.digitalocean.com/reference/api/reference/agent-platform-api/index.html.md) includes the endpoints for evaluation datasets, evaluation runs, scenario sets, the Common Scenario & Goal Library, and simulation runs.

## The DigitalOcean Command Line Client, doctl

[`doctl`](https://github.com/digitalocean/doctl) is the command-line interface for the DigitalOcean API. It supports most of the same actions available in the API and DigitalOcean Control Panel.

The [`doctl evaluation`](https://docs.digitalocean.com/reference/doctl/reference/evaluation/index.html.md) commands manage dataset evaluations and simulations:

- [Dataset](https://docs.digitalocean.com/reference/doctl/reference/evaluation/dataset/index.html.md): Create, list, and delete evaluation datasets.
- [Model run](https://docs.digitalocean.com/reference/doctl/reference/evaluation/model-run/index.html.md): Create, list, update, cancel, and delete evaluation runs, and get a download URL for run results.
- [Metric](https://docs.digitalocean.com/reference/doctl/reference/evaluation/metric/index.html.md) and [custom metric](https://docs.digitalocean.com/reference/doctl/reference/evaluation/custom-metric/index.html.md): List built-in evaluation metrics, and create, update, and delete custom metrics.
- [Preset](https://docs.digitalocean.com/reference/doctl/reference/evaluation/preset/index.html.md): List, get, and delete evaluation presets.
- [Scenario set](https://docs.digitalocean.com/reference/doctl/reference/evaluation/scenario-set/index.html.md): Create, generate, list, update, and delete scenario sets, and list their scenarios.
- [Scenario library](https://docs.digitalocean.com/reference/doctl/reference/evaluation/scenario-library/index.html.md): Browse library entries and their scenarios, and import an entry as a scenario set.
- [Simulation run](https://docs.digitalocean.com/reference/doctl/reference/evaluation/simulation-run/index.html.md): Create, list, update, cancel, and delete simulation runs, and get their episodes and trajectories.

See the [`doctl` documentation](https://docs.digitalocean.com/reference/doctl/reference/index.html.md) or run a command with `--help` for more information.

## The DigitalOcean MCP Server

The [DigitalOcean Evaluations MCP tools](https://docs.digitalocean.com/reference/mcp/evaluations-mcp-tools/index.html.md) let you use natural language prompts to manage evaluation datasets, run evaluations against agent deployments, and monitor results.