---
title: "Guides, the Promigence blog · Promigence"
description: "How to run a given agent, framework or workload in a sandbox, step by step."
url: "https://www.promigence.ai/blog/category/guides"
site: "Promigence"
---

# Guides

How to run a given agent, framework or workload in a sandbox, step by step.

- [All](https://www.promigence.ai/blog)
- [Guides](https://www.promigence.ai/blog/category/guides)

25 October 202616 min read

## [Best CI runner for GitHub Actions in 2026: hosted and self-hosted alternatives](https://www.promigence.ai/blog/best-ci-runner-for-github-actions-2026)

Promigence runners, GitHub-hosted runners, Blacksmith's and Depot's hosted runners, and self-hosted runners, compared on the workflow change, sizes, billing, spend control and upkeep.

- [guides](https://www.promigence.ai/blog/tags/guides)
- [ci runners](https://www.promigence.ai/blog/tags/ci-runners)
- [github actions](https://www.promigence.ai/blog/tags/github-actions)
- [blacksmith](https://www.promigence.ai/blog/tags/blacksmith)
- [depot](https://www.promigence.ai/blog/tags/depot)

19 October 20267 min read

## [Time to feedback: the delay that matters for coding agents](https://www.promigence.ai/blog/time-to-feedback-coding-agents)

A coding agent edits and checks its work dozens of times in one task. The arithmetic of why that loop outweighs how fast a sandbox starts, what keeps the loop short, and how to time your own.

- [guides](https://www.promigence.ai/blog/tags/guides)
- [coding agents](https://www.promigence.ai/blog/tags/coding-agents)
- [warm builds](https://www.promigence.ai/blog/tags/warm-builds)

19 October 20267 min read

## [What is Promigence? Sandboxes for AI agents, explained](https://www.promigence.ai/blog/what-is-promigence)

Promigence is a sandbox platform for AI agents: production-grade sandboxes, ready in milliseconds, fast at scale and billed by the second. What it is, what teams run on it, how using it looks, and how to get started.

- [guides](https://www.promigence.ai/blog/tags/guides)
- [agent sandboxes](https://www.promigence.ai/blog/tags/agent-sandboxes)
- [snapshots](https://www.promigence.ai/blog/tags/snapshots)
- [pricing](https://www.promigence.ai/blog/tags/pricing)

Follow Guides by [RSS](https://www.promigence.ai/blog/category/guides/rss.xml).

## Run your own workload on it, free

Send a repo and the command you run against it, whatever that is: an eval suite, an RL rollout, a CI job, a queue of coding tasks. We build the environment once, run it a thousand times, and send back the timings, the failures and an exact price. Free, once, on your real workload.

- 1,000 runs of your own command, on your own repo
- What each one cost in wall clock, and anything that failed
- Whether a failure was your code or the environment
- An exact price for your real volume

Not ready to hand over a repo? Read the [quickstart](https://www.promigence.ai/docs/quickstart) or [check the numbers first](https://www.promigence.ai/#numbers).