> For the complete documentation index, see [llms.txt](/llms.txt).
> A full single-fetch corpus is available at [llms-full.txt](/llms-full.txt).
---
title: Run with AGNT5
description: Deploy workers, trigger production runs, inspect traces, stream logs, and watch runtime metrics.
last_verified: 2026-10-09
---

Run covers the production lifecycle for deployed AGNT5 workers. You **[deploy](/docs/run/deploying.md)** a worker into an **[environment](/docs/run/environments.md)**, trigger workflows and agents, and use the recorded runs, traces, logs, and metrics to operate production behavior from Studio or the **[CLI](/docs/cli/inspect.md)**.

## Run guides


  - [Deploying](/docs/run/deploying) — Package your worker, deploy it to a managed environment, and verify the deployment.
  - [Environments](/docs/run/environments) — Separate preview, staging, and production, promote a verified deployment, and roll back.
  - [Streaming output](/docs/run/streaming-output) — Consume live output as it arrives and read the durable result when the run ends.
  - [Serverless endpoints](/docs/run/serverless-endpoints) — Run AGNT5 components on Cloudflare Workers, Vercel, Cloud Run, and other HTTP hosts.
  - [Operate serverless endpoints](/docs/run/operate-serverless-endpoints) — Verify, activate, disable, recover, and roll back provider deployments.
  - [Serverless support matrix](/docs/run/serverless-support-matrix) — Compare serverless runtimes, hosting targets, and framework adapters.
  - [Serverless protocol](/docs/run/serverless-protocol) — The signed manifest, invoke, checkpoint, and result contract between AGNT5 and an endpoint.
  - [Serverless SDK reference](/docs/run/serverless-sdk-reference) — Define serverless components, durable steps, and suspension in Python, TypeScript, and Go.


Outcome: your workflows and agents run outside your laptop, and every execution is observable.

## Next steps



After your code runs locally, take it to production and watch how it behaves.

- [Deploy a worker](/docs/run/deploying): Ship your worker to an environment and verify it.
- [Stream output](/docs/run/streaming-output): Show live output to your users while a run is in progress.
- [Measure quality](/docs/improve/overview): Turn production runs into datasets and experiments.