Ecosystem
Vercel Sandbox integrates with the agent frameworks, model SDKs, and coding agents you already use. Run a coding agent or give your agent a code execution tool inside an isolated Firecracker microVM.
The LangChain, OpenAI SDK, and Anthropic SDK guides define a run_code
tool backed by
sandbox.runCommand(),
pass it to the framework's tool-calling loop, and return the command output
to the model. Model requests route through AI Gateway,
while Vercel credentials authenticate the sandbox.
TanStack AI uses a dedicated Sandbox provider to run coding agents with workspace configuration and policies.
| Integration | Type | Use case |
|---|---|---|
| LangChain | Agent framework | Agents with a sandboxed code execution tool |
| OpenAI SDK | Model SDK | Function calling through the Responses API |
| Anthropic SDK | Model SDK | Tool use through the Messages API |
| AI SDK | AI SDK | Tool calling with the AI SDK |
| TanStack AI | Agent framework | Coding agents with sandboxed workspaces |
Build a v0-style app builder that turns a prompt into a Next.js app with a live preview. Users can refine the app through follow-up messages.
OpenAI Agents API manages the agent session, while Vercel Sandbox provides an isolated workspace to edit files and run the app. These integrations use the provider's API credentials.
| Integration | Type | Use case |
|---|---|---|
| OpenAI Agents API | Hosted agent API | OpenAI-managed sessions with Sandbox execution and persistent files |
These integrations give an agent a filesystem view of a content or object store, so it reads and writes through standard file operations instead of calling a storage API. The files live in the provider, not the sandbox. That means the provider's permissions decide what the agent can open, and the agent's work is still there after the sandbox is deleted.
| Integration | Type | Use case |
|---|---|---|
| Box Mount | Content platform | Agents reading and writing Box content as local files |
| Integration | Type | Use case |
|---|---|---|
| Cursor | Coding agent | Cursor Cloud Agents on Self-Hosted Machines |
| Devin | Coding agent | Devin Outposts sessions in Sandbox microVMs |
| Herdr | Agent manager | Terminal coding agents in persistent Sandboxes |
| Hermes | Coding agent | Hermes terminal commands in Sandbox microVMs |
| Integration | Type | Use case |
|---|---|---|
| Harbor | Eval harness | Terminal-Bench and benchmark trials in Sandbox microVMs |
Was this helpful?