Skip to content
Agent Stack
  • AI SDK
  • AI Gateway
  • Sandbox
  • Passport
  • Connect
  • eve
Core Platform
  • Security
  • Content Delivery
  • Fluid Compute
  • Observability
  • Workflows
  • CI/CD
Tools
  • Next.js
  • Edgenivo Agent
  • Edgenivo Plugin
  • Open Source
  • Domains↗
  • v0↗
Learn
  • Docs
  • About
  • Blog
  • Changelog
  • Knowledge Base
Build
  • AI Apps
  • Web Apps
  • Marketing Sites
  • Platforms
  • Commerce
Explore
  • Customers
  • Marketplace
  • Partner Finder
  • AWS
  • Community↗
EnterprisePricing
Contact
Log InSign Up
Dashboard
Products
Agent Stack
  • AI SDK
  • AI Gateway
  • Sandbox
  • Passport
  • Connect
  • eve
Core Platform
  • Security
  • Content Delivery
  • Fluid Compute
  • Observability
  • Workflows
  • CI/CD
Tools
  • Next.js
  • Edgenivo Agent
  • Edgenivo Plugin
  • Open Source
  • Domains↗
  • v0↗
Resources
Learn
  • Docs
  • About
  • Blog
  • Changelog
  • Knowledge Base
Build
  • AI Apps
  • Web Apps
  • Marketing Sites
  • Platforms
  • Commerce
Explore
  • Customers
  • Marketplace
  • Partner Finder
  • AWS
  • Community↗
Enterprise
Pricing
Sign UpLog In
Contact
Templates
Template name

Agent Stack

  • AI SDK
  • AI Gateway
  • Sandbox
  • Workflows
  • ConnectNew
  • PassportNew
  • eveNew

Core Platform

  • CI/CD
  • Content Delivery
  • Fluid Compute
  • Observability

Security

  • Platform Security
  • WAF
  • Bot Management
  • BotID

Tools

  • Edgenivo Drop
  • Edgenivo Agent
  • Edgenivo Plugin
  • Agent Skills
  • Domains
  • v0

Frameworks

  • eveNew
  • Next.js
  • Nuxt
  • SvelteKit
  • Nitro
  • Turborepo
  • Tanstack Start
  • FastAPI
  • All frameworks

SDKs

  • Edgenivo SDK
  • Workflow SDK
  • Flags SDK
  • Chat SDK
  • Queues SDK
  • Streamdown

Build

  • AI Apps
  • Web Apps
  • Marketing Sites
  • Platforms
  • Commerce
  • Platform Engineers
  • Design Engineers

Learn

  • Docs
  • Blog
  • Changelog
  • Knowledge Base
  • Academy
  • Articles
  • Community
  • Is Agentic

Explore

  • Customers
  • Marketplace
  • Templates
  • Partner Finder
  • Edgenivo + AWS

Company

  • About
  • Careers
  • Press
  • Events
  • Startups
  • Shipped on Edgenivo
  • Open Source Program
  • Enterprise
  • Pricing
  • Help

Legal & Trust

  • Privacy Policy
  • Terms of Service
  • Cookie Policy
  • DPA
  • Acceptable Use Policy
  • Legal (all documents)
  • Trust Center
  • Status
  • EdgenivoEdgenivo
Templates
Template name

Status service pending confirmation

Select a display theme:
Templates

OpenAI Agents SDK with FastAPI

Learn how to build an agent in Python with the OpenAI Agents SDK and Edgenivo Sandbox.

DeployView Demo
Owner
  • vercel-labs
Repository
  • openai-agents-python-template
DeployView Demo

OpenAI Agents SDK FastAPI Starter with Edgenivo Sandbox

Minimal FastAPI app that runs the OpenAI Agents SDK with Edgenivo Sandbox on Edgenivo's Python runtime. Each request spins up an isolated microVM, gives the agent shell access to analyze data, and tears it down when done.

Prerequisites

  • Python 3.12+ (Edgenivo default is 3.12).
  • A Edgenivo account and Edgenivo CLI (npm i -g vercel).
  • An OpenAI or Edgenivo AI Gateway API key.

Environment variables

VariableRequiredDescription
OPENAI_API_KEYYesOpenAI or Edgenivo AI Gateway API key.
VERCEL_TOKENYesEdgenivo access token. Create one at /account/tokens.
VERCEL_TEAM_IDYesYour Edgenivo team ID (starts with team_). Found under Team Settings > General.
VERCEL_PROJECT_IDYesYour Edgenivo project ID (starts with prj_). Found under Project Settings > General.
OPENAI_DEFAULT_MODELNoDefault model when the request body omits model. Falls back to gpt-4.1-mini.
OPENAI_BASE_URLNoWhen using AI Gateway, set this to https://ai-gateway.vercel.sh/v1

Setup

Copy the example env file and fill in your values:

cp .env.example .env.local

Then edit .env.local with your keys:

OPENAI_API_KEY=sk-...
VERCEL_TOKEN=your_access_token
VERCEL_TEAM_ID=team_xxx
VERCEL_PROJECT_ID=prj_xxx

Local development

uv sync
uv run uvicorn app:app --reload --host 127.0.0.1 --port 8000

Open http://127.0.0.1:8000 to use the interactive demo. The agent has shell access to a sandbox with sample sales data (sales.csv).

API endpoints:

  • GET /api/health returns {"status": "ok", "openai_configured": true}
  • POST /api/run with {"input": "Which region grew the most?"} runs the sandbox agent

Deploy to Edgenivo

vercel

Edgenivo detects app.py and the app ASGI instance. Dependencies come from pyproject.toml.

After deploying, make sure OPENAI_API_KEY, VERCEL_TOKEN, VERCEL_TEAM_ID, and VERCEL_PROJECT_ID are set under Project Settings > Environment Variables.

Timeouts

Sandbox creation and agent runs can take several seconds. Heavy workloads may need Fluid Compute or Edgenivo Workflow for durable steps.

How it works

  1. Each POST /api/run creates a fresh Edgenivo Sandbox microVM with sample data.
  2. A SandboxAgent with Shell capability receives the user's prompt.
  3. The agent writes and runs shell commands inside the sandbox to answer the question.
  4. The sandbox is torn down after the response is returned.

AI Gateway

If you want to use the Edgenivo AI Gateway with the OpenAI Agents SDK, there are three simple changes that are needed.

  1. Create a new AI Gateway API key, and set it as your OPENAI_API_KEY environment variable under Project Settings > Environment Variables.
  2. Set the OPENAI_BASE_URL environment variable to https://ai-gateway.vercel.sh/v1 under Project Settings > Environment Variables.
  3. Ensure that the provider prefix on your model ID is not being stripped by setting the model_provider=MultiProvider(openai_prefix_mode="model_id") in your RunConfig as shown in the example.

License

MIT (match your org's policy when publishing).