---
title: "eve Agent with Python"
description: "Connect an eve agent and Next.js chat to FastAPI for Python analysis and shared tasks, with Neon storage in one Vercel project."
url: "https://vercel.com/templates/eve/eve-agent-with-python"
links:
  repository: "https://github.com/vercel-labs/eve-agent-python-template"
  deploy: "https://vercel.com/new/clone?demo-title=eve+Agent+with+Python&demo-description=Connect+an+eve+agent+and+Next.js+chat+to+FastAPI+for+Python+analysis+and+shared+tasks%2C+with+Neon+storage+in+one+Vercel+project.&project-name=eve-agent-python&repository-name=eve-agent-python&repository-url=https%3A%2F%2Fgithub.com%2Fvercel-labs%2Feve-agent-python-template&env=EVE_CHAT_PASSWORD&envDescription=The+password+visitors+enter+before+they+can+use+the+app+and+chat+with+the+agent.&envLink=https%3A%2F%2Fgithub.com%2Fvercel-labs%2Feve-agent-python-template%23getting-started&products=%5B%7B%22type%22%3A%22integration%22%2C%22protocol%22%3A%22storage%22%2C%22productSlug%22%3A%22neon%22%2C%22integrationSlug%22%3A%22neon%22%7D%5D&demo-image=%2F%2Fimages.ctfassets.net%2Fe5382hct74si%2F2JipXpxaYyBFDSqfw1Lmjb%2F10378b8548219cc205731d1c5c4f3e7c%2Feve-agent-with-python-3x.png"
---

# eve Agent with Python

Connect an eve agent and Next.js chat to FastAPI for Python analysis and shared tasks, with Neon storage in one Vercel project.

# eve Agent with Python

eve Agent with Python connects a Next.js chat interface to a FastAPI backend for calculations and shared application data. Built for developers adding Python business logic to React apps, it includes an eve AI agent that manages tasks through chat and shows Python analysis results as cards.

## What can you build with this template?

Start with the included task manager. Ask the agent to add a task, then open the Tasks page to see it. Both interfaces call the same FastAPI endpoints to create, edit, complete, and delete records. The agent's deletion tool pauses for approval before removing a task.

Two analysis tools show how to return Python results to the conversation. The statistics tool computes mean, median, quartiles, and other measures. Text analysis returns word counts, estimated reading time, readability, and frequent words. FastAPI validates requests and runs the calculations, while the chat displays the returned data in dedicated cards.

For other responses, json-render turns structured output into tables, metrics, and components from a developer-defined catalog. Component props are validated before rendering. Follow-up forms accept choices or typed answers, and conversations resume after a page reload.

## Why use this template?

With the agent, backend, and interface already connected, you can extend the task manager into an internal operations app. The shared endpoints let artificial intelligence agents and people using conventional pages work with the same records.

You can build AI agents around your own Python functions by adding backend endpoints and TypeScript tools that call them. The [repository](https://github.com/vercel-labs/eve-agent-python-template) also shows where to customize result cards and the component catalog. If you're choosing the best AI agent starter for a Python-backed application, look for this separation between conversational behavior, backend logic, and presentation.

## Why use eve on Vercel?

[eve is an open-source AI agent framework](https://vercel.com/blog/introducing-eve). In this template, it runs the TypeScript agent and its tools, while FastAPI handles Python requests. [Vercel Services](https://vercel.com/docs/services) deploys the agent, Next.js frontend, and FastAPI backend together in one project.

The backend has no public route. Internal service bindings supply its URL to the agent and Next.js server, so browser requests go through the web app. This keeps the Python service part of the application without requiring a separate public API deployment.

## What should you know before deploying?

Follow the [repository setup guide](https://github.com/vercel-labs/eve-agent-python-template#getting-started) to configure Neon and set `EVE_CHAT_PASSWORD`. SQLModel defines the task data model, and Neon Postgres stores tasks on Vercel. Local development uses SQLite when `DATABASE_URL` is unset; deployed task endpoints require the database connection.

Everyone with the password shares one identity and task list, and can open conversation links shared with them. Use this access model only for people who may see each other's chats and tasks. Separate private accounts require your own authentication plus conversation ownership checks and user-scoped task queries. The agent's deletion approval does not establish an approval policy for every UI action.

## Frequently asked questions

### Is the eve agent written in Python?

No. The agent's tools are TypeScript files that call FastAPI endpoints. Python implements the backend calculations and data operations, and Next.js provides the React interface.

### What can the Python backend analyze, and where are tasks saved?

It calculates numerical statistics and text measures such as reading time and word frequency. Task records persist in Neon Postgres on Vercel through SQLModel, with SQLite available as the local fallback when no database URL is configured.

### Can browsers call the FastAPI service directly?

The supplied deployment exposes no public route to FastAPI. The web app forwards browser requests, while the agent and Next.js server reach the backend through internal service bindings.

### How does the agent generate UI components?

The agent produces a json-render specification using the catalog you define. The template checks component props before rendering them with the registered React components. Extend both the catalog and its renderers to add new UI options.

### Does the password give each person a private workspace?

No. Password holders have equal access to the shared tasks and can open each other's conversation links. To isolate users, replace shared sign-in and enforce conversation and task ownership.

## Related resources

- [eve Next.js integration guide](https://eve.dev/docs/guides/frontend/nextjs)
- [Vercel Services documentation](https://vercel.com/docs/services)

## Related Templates

### [eve TanStack Agent](https://vercel.com/templates/eve/eve-tanstack-agent)

Build a TanStack Start chat with eve, interactive cards, sortable tables, follow-up forms, and conversations with shareable URLs.

### [eve Data Analyst](https://vercel.com/templates/eve/eve-data-analyst)

Ask data questions in Slack and get Postgres-backed answers with charts and SQL evidence. Configure metrics, roles, and query approvals.

### [eve with Vercel Drives](https://vercel.com/templates/eve/eve-vercel-drives-starter)

Build an eve agent for shared team files on Vercel Drives, with read-only session snapshots and a separate tool for saving files.

## Related Integrations

### [Neon](https://vercel.com/marketplace/neon)

The backend for apps and agents, built to scale on Lakebase Postgres.

---

## Additional documentation

The [Vercel Knowledge Base](https://vercel.com/kb) contains guides and answers to common questions about Vercel, including deployment, framework configuration, domains, caching, and troubleshooting.
