#6262nd of 118 in Agent frameworks and SDKs
eve
TypeScript framework for durable agents defined as files in an agent/ directory
- Stars
- 5.9k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Oct 2026
- Language
- TypeScript
22 on 16 Jun 20265,856 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Vercel is hiring: 87 open roles
Overview
eve is a TypeScript framework where an agent is defined by a conventional directory: instructions.md, typed tools, on-demand skills, message channels (HTTP, Slack, Discord) and cron schedules. It scaffolds with `npx eve@latest init` and picks models by AI Gateway model ID. Web Chat can be added with Vercel sign-in, and `eve deploy` publishes the agent.
Who it is for: TypeScript developers building scheduled, chat-connected agents
Strengths
- Agent layout is plain files: instructions.md, tools/, skills/, channels/, schedules/
- Tools are typed functions defined with defineTool and Zod input schemas
- Built-in channels for HTTP, Slack and Discord, plus cron schedules
- Package ships its docs in node_modules/eve/docs for coding agents
Weaknesses
- Models are selected by AI Gateway model ID, tying setup to that gateway
- Web Chat sign-in and deploy flow are built around Vercel projects
- No Docker or compose files provided for self-hosting
- Local model support is not described in the README
What it needs
- no GPU
- Needs Node.js and npm, Vercel AI Gateway, Vercel project (for Web Chat and deploy)
- Models: AI Gateway model IDs (e.g. openai/gpt-5.6-terra)
Quick answers
Is eve open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Does eve need a GPU?
No, it runs without one.
Also in Agent frameworks and SDKs
See all 118| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 95 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 91 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 91 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitSDK for adding agent chat, generative UI and shared state to apps | 90 out of 100 |