nerve

Nerve 是一个智能体开发工具包（ADK），专为技术用户设计，通过 YAML 和 CLI 构建、运行、评估和编排基于大语言模型（LLM）的智能体。它支持声明式智能体定义、内置工具与扩展性、原生 MCP 客户端与服务器支持、评估模式、工作流组合以及 LLM 无关性。Nerve 允许用户通过简洁的 YAML 文件定义智能体，并使用多种工具和模型实现自动化。其核心功能包括智能体团队协作、深度编排、基准测试和多步骤自动化。Nerve 还支持通过 LiteLLM 切换多种模型，并提供了详细的文档和示例，帮助用户快速上手。

article

README

`nerve`

The Simple Agent Development Kit

Join the project community on our server!

Nerve is a simple yet powerful Agent Development Kit (ADK) to build, run, evaluate, and orchestrate LLM-based agents using just YAML and a CLI. It’s designed for technical users who want programmable, auditable, and reproducible automation using large language models.

Key Features

📝 Declarative Agents

Define agents using a clean YAML format: system prompt, task, tools, and variables — all in one file.

🔧 Built-in Tools & Extensibility

Use shell commands, Python functions, or remote tools to power your agents. Tools are fully typed and annotated.

🌐 Native MCP Support (Client & Server)

Nerve is the first framework to let you define MCP servers in YAML — and act as both client and server, enabling agent teams and deep orchestration.

📊 Evaluation Mode

Benchmark your agents with YAML, Parquet, or folder-based test cases. Run reproducible tests, log structured outputs, and track regression or progress.

🔁 Workflows

Compose agents into simple, linear pipelines to create multi-step automations with shared context.

🧪 LLM-Agnostic

Built on LiteLLM, Nerve supports OpenAI, Anthropic, Ollama, and dozens more — switch models in one line.

Quick Start

# 🖥️ install the project with:
pip install nerve-adk

# ⬇️ download and install an agent from a github repo with:
nerve install evilsocket/changelog

# 💡 or create an agent with a guided procedure:
nerve create new-agent

# 🚀 go!
nerve run new-agent

Read the documentation and the examples for more.

Contributing

We welcome contributions! Check out our contributing guidelines to get started and join our Discord community for help and discussion.

License

Nerve is released under the GPL 3 license.