--- title: Vercel Sandbox description: The compute primitive designed to safely run untrusted or user-generated code on Vercel. url: "https://vercel.com/kb/vercel-sandbox" published: 2026-06-15 last_updated: 2026-09-11 install_vercel_plugin: npx plugins add vercel/vercel-plugin --- # Vercel Sandbox The compute primitive designed to safely run untrusted or user-generated code on Vercel. ## Using Vercel Sandbox - [Build a Claude Managed Agent with Vercel Sandbox](/kb/guide/run-claude-managed-agent-tools-with-vercel-sandbox): Build a Claude Managed Agent with Vercel Sandbox: each session runs in a fresh microVM with credential brokering and a webhook-driven control plane on Vercel. - [How Vercel Sandbox duration and persistence work](/kb/guide/vercel-sandbox-duration-and-persistence): Session duration and persistence are two separate controls in Vercel Sandbox. The timeout option keeps a single run alive for up to 24 hours, while snapshots and persistent sandboxes carry state across runs for long-running agents and pipelines. - [Run Python code securely with AI SDK and Vercel Sandbox](/kb/guide/python-ai-sdk-vercel-sandbox): Add an \`executeCode\` tool to your AI SDK agent with the \`ai-sdk-tool-code-execution\` package to run Python 3.13 inside an isolated Vercel Sandbox. - [Investigate GitHub issues with \`HarnessAgent\` and Vercel Sandbox](/kb/guide/sandboxed-coding-agent-with-harnessagent): Use HarnessAgent to run coding agent harnesses like Claude Code or Codex against untrusted code inside Vercel Sandbox. Reproduce GitHub issues with the agent harness and generate a report for the maintainer. - [Building an agent with OpenAI Agents SDK and Vercel Sandbox](/kb/guide/building-an-agent-with-openai-agents-sdk-and-vercel-sandbox): Learn how to build an agent with with OpenAI Agents SDK and Vercel Sandbox - [Running OpenClaw in Vercel Sandbox](/kb/guide/running-openclaw-in-vercel-sandbox): This guide walks you through setting up OpenClaw inside a Vercel Sandbox and configuring the WhatsApp channel. - [Running OpenCode securely with the Vercel Sandbox](/kb/guide/running-opencode-securely-with-the-vercel-sandbox): Run OpenCode in an isolated Vercel Sandbox MicroVM with controlled egress, using the SDK to restrict network access so the AI agent can only reach approved LLM endpoints. - [Safely running AI generated code in your Next.js application](/kb/guide/running-ai-generated-code-sandbox): How to execute untrusted, AI‑generated code from a Next.js app using Vercel Sandbox, an isolated, ephemeral environment. ## Build with OpenAI Agents API - [Build a v0-style app builder with OpenAI Agents API and Vercel Sandbox](/kb/guide/v0-clone-openai-agents-vercel-sandbox): Build a chat-to-app workspace with live Next.js previews, follow-up edits, and saved projects using the OpenAI Agents API preview, Vercel Sandbox, and Queues. - [Build an agentic app in FastAPI with OpenAI Agents API and Vercel Sandbox](/kb/guide/fastapi-openai-agents-api-vercel-sandbox): Learn how to build a repository Q\&A app using FastAPI, OpenAI Agents API, and Vercel Sandbox, with isolated code inspection, streamed progress, downloadable evidence, and deployment to Vercel. ## Build an agent with eve - [Build a software factory with eve](/kb/guide/eve-software-factory): Foreman, a software factory built on eve. It turns GitHub and Linear work items into reviewed draft pull requests through four agent stations, triaging issues and fixing its own red CI along the way. - [Build an incident response sre agent with eve](/kb/guide/eve-incident-sre-agent): sre is an incident response agent for Slack. It investigates production issues using a hypothesis-driven approach and returns verifiable evidence from connected sources. - [How to automatically approve tool calls in eve with Jev](/kb/guide/auto-approve-tool-calls-eve-jev): Use Jev to review tool calls in eve, allow routine actions, and request human approval when needed. Configure the policy and test its failure paths. - [Run a marketing team from Slack with eve](/kb/guide/marketing-team-eve): Team of five marketing agents built on eve. The lead routes work to specialists that write long-form into Notion, queue social posts in Typefully, run email in Resend, and audit for SEO, with approvals for anything irreversible. - [How to build a browser agent that works behind a login](/kb/guide/build-a-browser-agent): Build a browser agent with eve, Vercel Connect, and KERNEL managed auth that signs a user in through a human-in-the-loop handoff and completes tasks behind a login without exposing credentials. - [Give your eve agent an email inbox with Resend](/kb/guide/eve-agent-with-resend): Connect an eve agent to email using the Chat SDK channel and the Resend adapter. Handle inbound email, threaded replies, attachments, and proactive sends. - [Manage your Sanity project from Slack with eve](/kb/guide/eve-sanity-copilot): A Slack-based Sanity copilot built on eve. It queries and edits content with GROQ, shapes schemas, manages releases, and drafts long-form pieces into Notion. ## Build with a eve Template - [Data analyst](https://vercel.com/templates/eve/eve-data-analyst): Slack-based data analyst built on eve that answers questions about your business from a Postgres database. - [Contentful assistant](https://vercel.com/templates/eve/eve-contentful-agent): Slack-based Contentful assistant with Vercel Drives providing a searchable copy of your Contentful content. - [Marketing team](https://vercel.com/templates/eve/eve-marketing-team): A team of marketing agents: a lead routes work to specialists for positioning, long-form content, social, SEO, and email, publishing through Notion, Typefully, and Resend. - [Software factory](https://vercel.com/templates/eve/eve-software-factory): Software factory built on eve: AI agents work each stage of the development loop, and people make the judgment calls. - [Incident response](https://vercel.com/templates/eve/eve-incident-response-agent): sre investigates production issues using a hypothesis-driven approach and outputs verifiable evidence from sources. - [GitHub maintainer](https://vercel.com/templates/eve/eve-github-maintainer): Kody, a personal GitHub maintainer agent that emails you a weekly digest of open issues, acts on your email replies, summarizes new pull requests, answers @mentions, and works delegated Linear issues. - [Personal agent](https://vercel.com/templates/nuxt/eve-personal-agent): A durable AI assistant with long-term memory. Chat on the web or Slack, query Linear, and pick up where you left off. - [Vercel Drives Starter](https://vercel.com/templates/eve/eve-vercel-drives-starter): An eve agent that reads and saves your team's files on a Vercel Drive. - [Memory](https://vercel.com/templates/eve/eve-memory-agent): A Slack agent built on eve that remembers what your team tells it. Personal preferences and channel conventions are stored separately in Vercel Blob and recalled before every reply. - [TanStack Start agent](https://vercel.com/templates/eve/eve-tanstack-agent): Web chat built on eve and TanStack Start, where the agent can respond with generative UI like tables and dashboards. ## All Vercel Sandbox guides - [How to use Vercel Drives with an eve agent](/kb/guide/vercel-drives-eve-agent): Mount Vercel Drives in an eve agent's sandbox to give each user a persistent workspace and share read-only reference data across sessions. - [Build a v0-style app builder with OpenAI Agents API and Vercel Sandbox](/kb/guide/v0-clone-openai-agents-vercel-sandbox): Build a chat-to-app workspace with live Next.js previews, follow-up edits, and saved projects using the OpenAI Agents API preview, Vercel Sandbox, and Queues. - [Build an agentic app in FastAPI with OpenAI Agents API and Vercel Sandbox](/kb/guide/fastapi-openai-agents-api-vercel-sandbox): Learn how to build a repository Q\&A app using FastAPI, OpenAI Agents API, and Vercel Sandbox, with isolated code inspection, streamed progress, downloadable evidence, and deployment to Vercel. - [Build an agent with OpenAI Agents API on Vercel](/kb/guide/openai-agents-api-vercel): Build and deploy an agent with OpenAI Agents API, Vercel Functions, Queues, and Sandbox for isolated code execution. - [Using TanStack AI with Vercel Sandbox](/kb/guide/tanstack-ai-vercel-sandbox): Run coding agents in isolated Vercel Sandbox microVMs with the @tanstack/ai-sandbox-vercel provider, with durable resume-by-id, port previews, and workspace policies. - [Run Cursor Cloud Agents on Vercel Sandbox](/kb/guide/cursor-vercel-sandbox): Learn how to run Cursor Cloud Agents on Vercel Sandbox with BYOM worker pools, durable workflows, isolated microVMs, and scale-to-zero compute. - [How to run Herdr coding agents in isolated Vercel Sandboxes](/kb/guide/run-herdr-coding-agents-isolated-vercel-sandboxes): Install the vercel.sandbox plugin for Herdr, approve an upload manifest, run each AI coding agent in its own isolated Vercel Sandbox microVM, and apply its changes back to your local worktree. - [How to run Devin Outposts sessions on Vercel](/kb/guide/devin-outposts-vercel-sandbox): Run Devin Outposts sessions in isolated Vercel Sandbox microVMs. One browser authorization to set up, then sessions start from anywhere, even your phone. - [How to test a container image in Vercel Sandbox before deploying](/kb/guide/test-container-image-vercel-sandbox): Validate a container image before deploying by booting it as a custom Sandbox image from Vercel Container Registry (VCR) or running Docker inside a Vercel Sandbox. - [Manage your Sanity project from Slack with eve](/kb/guide/eve-sanity-copilot): A Slack-based Sanity copilot built on eve. It queries and edits content with GROQ, shapes schemas, manages releases, and drafts long-form pieces into Notion. - [How to process user-uploaded files with Vercel Sandbox and Vercel Blob](/kb/guide/user-uploaded-files-vercel-sandbox-and-blob): Learn how to upload files to Vercel Blob, process them safely with FFmpeg in an isolated Vercel Sandbox, and store the results back in Blob. - [How to use Vercel Container Registry](/kb/guide/how-to-use-vercel-container-registry): Push, store, and pull OCI container images with Vercel Container Registry, then deploy them to Vercel Functions and Vercel Sandbox. - [How Vercel Sandbox duration and persistence work](/kb/guide/vercel-sandbox-duration-and-persistence): Session duration and persistence are two separate controls in Vercel Sandbox. The timeout option keeps a single run alive for up to 24 hours, while snapshots and persistent sandboxes carry state across runs for long-running agents and pipelines. - [Run Python code securely with AI SDK and Vercel Sandbox](/kb/guide/python-ai-sdk-vercel-sandbox): Add an \`executeCode\` tool to your AI SDK agent with the \`ai-sdk-tool-code-execution\` package to run Python 3.13 inside an isolated Vercel Sandbox. - [Investigate GitHub issues with \`HarnessAgent\` and Vercel Sandbox](/kb/guide/sandboxed-coding-agent-with-harnessagent): Use HarnessAgent to run coding agent harnesses like Claude Code or Codex against untrusted code inside Vercel Sandbox. Reproduce GitHub issues with the agent harness and generate a report for the maintainer. - [Draft content in your voice from Slack with eve](/kb/guide/eve-content-agent): Deploy the eve content agent template, a Slack bot that drafts blog posts, LinkedIn posts, release notes, and newsletters in your house voice from Notion, using Vercel Connect, Vercel Blob, and AI Gateway. - [Building an agent with OpenAI Agents SDK and Vercel Sandbox](/kb/guide/building-an-agent-with-openai-agents-sdk-and-vercel-sandbox): Learn how to build an agent with with OpenAI Agents SDK and Vercel Sandbox - [Build a Claude Managed Agent with Vercel Sandbox](/kb/guide/run-claude-managed-agent-tools-with-vercel-sandbox): Build a Claude Managed Agent with Vercel Sandbox: each session runs in a fresh microVM with credential brokering and a webhook-driven control plane on Vercel. ## Explore other topics in the Vercel Knowledge Base - [Python](/kb/python): Build, deploy, and scale Python on Vercel. Run FastAPI, Flask, Django, or any ASGI or WSGI app on Vercel Functions with zero configuration, and find guides on background jobs, storage, AI workloads, containers, and migrating from other platforms. - [Enterprise Apps and Agents](/kb/enterprise): Explore how enterprises develop, deploy, and govern internal apps and AI agents. Learn about citizen development, shadow AI, access controls, and agent auditing, with practical guidance for turning employee-built tools into supported business applications. - [Jev from TypeSafe AI](/kb/jev-from-typesafe-ai): Jev is TypeSafe AI's System One model for fast, typed decisions. It evaluates supplied context and returns choices, scores, and probabilities that applications can use to classify requests, route tasks, and assess proposed actions without generating prose. - [FDE Field Notes](/kb/fde): Insights from our work with customers. Explore the patterns, diagnostics, and recommendations Vercel’s Forward Deployed Engineers use in production.