nerve
★ 1.3k·GPL3·Updated 2025-08-23
Nerve is an Agent Development Kit (ADK) that allows technical users to build, run, evaluate, and orchestrate LLM-based agents using declarative YAML configurations and a command-line interface. It emphasizes programmable, auditable, and reproducible automation with support for multi-agent communication and various LLMs.
#Agent Development Kit#LLM Agents#YAML Configuration#CLI Tool#Multi-Agent Systems
01
Features
01Declarative Agents: Define agents using a clean YAML format including system prompt, task, tools, and variables.
02Built-in Tools & Extensibility: Utilize shell commands, Python functions, or remote tools, all fully typed and annotated.
03Native MCP Support (Client & Server): The first framework to define MCP servers in YAML, acting as both client and server for deep orchestration.
04Evaluation Mode: Benchmark agents with YAML, Parquet, or folder-based test cases for reproducible tests, structured outputs, and progress tracking.
05Workflows: Compose agents into simple, linear pipelines to create multi-step automations with shared context.
02
Compatibility
LiteLLM
Native Integration
Verified via docs
OpenAI
Supported
Verified via docs
Anthropic
Supported
Verified via docs
Ollama
Supported
Verified via docs
Multi-Agent Communication Protocol (MCP)
Native
Verified via docs
03
Quick start
1
$ pip install nerve-adk
04
Use cases
↳Automating complex tasks with LLMs (e.g., custom chatbots, data processing).
↳Developing and testing multi-agent systems and deep orchestrations.
↳Benchmarking and evaluating LLM agent performance and tracking regressions.
05
Alternatives
editor★ 25.1k
Open-source 3D architectural editor with a local CLI, MCP tools, and practical workflows for humans and AI agents.
capa★ 718
One capabilities.yaml wires skills, tools, rules, sub-agents, MCP servers, and plugins into Cursor, Claude Code, Codex, Windsurf, GitHub Copilot, and 30+ other AI coding agents
squirrelscan★ 299
The website QA tool for your coding agent. 300+ audit rules across SEO, performance, security, accessibility and agent experience.
Comments
Log in to leave a comment
No comments yet. Be the first!