# eve

A filesystem-first, Apache-2.0 framework for building durable backend AI agents that run on Vercel or your own infrastructure. eve is currently in beta.

This file helps AI agents understand and use this documentation site.

## Documentation Surfaces

- [Docs](https://localhost:80/docs): Documentation section: docs
- [Documentation index](https://localhost:80/llms.txt): Curated task-oriented map of the most useful documentation
- [Full documentation context](https://localhost:80/llms-full.txt): All configured documentation as Markdown
- [Semantic documentation sitemap](https://localhost:80/sitemap.md): High-level documentation map for agent navigation
- Canonical HTML pages use `/docs/...`, `/integrations/...`, and `/templates/...` URLs.
- Docs and integration pages expose alternate Markdown by appending `.md` or `.mdx` to their canonical HTML URL.
- Template pages are HTML discovery pages; use their setup prompt or source link instead of appending a Markdown extension.

## When to use eve

- Category: Agent framework
- Audience: developers building AI agents, teams running agents on any infrastructure
Common use cases:

- Create durable agents with filesystem conventions
- Add channels, tools, skills, sandboxes, hooks, and schedules
- Deploy agent workloads on Vercel or self-host them as Node services

## Important Links

- [Docs](https://localhost:80/docs)
- [Integrations](https://localhost:80/integrations)
- [Templates](https://localhost:80/templates)
- [GitHub repository](https://github.com/vercel/eve): Source repository

## Agent Instructions

- To create or extend an eve agent for the user, start from the Getting Started guide — get it as Markdown from /docs/getting-started.md (or via /llms.txt).
- Ask the user only for genuine decisions (name, model, channels, provider, deploy) and for browser/OAuth steps (vercel login, vercel link, vercel connect create slack); automate everything else.
- Verify setup with `eve info --json` and `eve channels list --json` before reporting success.
- Use /llms.txt as a concise task-oriented index and /sitemap.md as the exhaustive page map.
- Use /llms-full.txt only when you need the complete documentation corpus for offline indexing or a large context window.
- Fetch individual docs or integration pages with a .md or .mdx extension for focused page-level context. Template pages are HTML discovery pages and do not expose this alternate Markdown route.
- Treat eve.dev as framework documentation, not a shared API, authorization server, MCP server, or A2A server. Every deployed eve app exposes its own /eve/v1 routes and authentication policy; external API, OpenAPI, and MCP URLs in the docs may describe third-party connections or examples.
- Do not assume API, authentication, OpenAPI, or MCP support unless it is listed in this file.

## Templates

- [Chat](https://localhost:80/templates/eve-chat-template): A persisted Next.js chat template for eve, built with shadcn/ui, Tailwind CSS, Streamdown, Better Auth, Drizzle, Neon, and Upstash Redis.
- [LLM council](https://localhost:80/templates/eve-llm-council-template): A Next.js LLM council that sends one prompt to four models in parallel, streams their answers, and asks a judge model for a concise answer with per-model agreement scores.
- [Design](https://localhost:80/templates/eve-design-template): A Slack design collaborator that answers from a reviewed, versioned corpus of your organization's approved design guidance.
- [Slack](https://localhost:80/templates/eve-slack-agent): A Slack agent template with webhook handling, Vercel Connect, a starter agent, and an example tool ready to deploy on Vercel.
- [GitHub maintainer](https://localhost:80/templates/kody-eve-template): 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.
- [Software factory](https://localhost:80/templates/eve-software-factory-template): Foreman, a software factory that takes tasks from GitHub and Linear, runs each through classifier, analyst, implementer, and reviewer stations, and delivers a reviewed draft pull request on your repository.
- [Incident response](https://localhost:80/templates/eve-sre-agent-template): sre, an incident response agent that investigates alerts from Slack or webhooks using live Datadog, GitHub, and Vercel evidence, with read-only access by default.
- [Marketing team](https://localhost:80/templates/marketing-team-eve-template): 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.
- [Sanity copilot](https://localhost:80/templates/sanity-copilot-eve-template): A Slack-based Sanity copilot that queries and edits content with GROQ, inspects and shapes schemas, manages drafts and releases, and drafts long-form pieces into Notion.
- [Social media](https://localhost:80/templates/typefully-eve-template): A Slack-based social media agent that drafts posts and threads for X, LinkedIn, Threads, Bluesky, and Mastodon through Typefully, manages the publishing queue, and pulls briefs from Notion.
- [Weather](https://localhost:80/templates/weather-agent-fixture): A small representative eve app with agent config, instructions, a typed weather tool, and a markdown skill.