Search

More Filters

Show More
Add Listing
Go Back

nerve

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

Nerve is a lightweight Agent Development Kit (ADK) for building, running, evaluating, and orchestrating LLM-based agents using declarative YAML files and a command line interface. It is aimed at technical users who want programmable, auditable, and reproducible automation driven by large language models. The project centralizes agent configuration into single YAML files that include system prompts, tasks, tools, and variables. Nerve supports a variety of tool types including shell commands, Python functions, and remote tools, and is designed to be LLM-agnostic through an integration with LiteLLM providers. The ADK also includes evaluation features, workflow composition for multi-step automations, and native MCP client and server support so teams of agents can be orchestrated. Documentation, examples, a PyPI package, Docker images, and a community hub are provided for users.

Links

Stars
1131
App URL
https://github.com/evilsocket/nerve
Github Repository
https://github.com/evilsocket/nerve/blob/main/README.md

Categorization

App Details

Features
Declarative agent definitions in YAML allow system prompts, task statements, tools, and variables to live in one file. Built-in and extensible tools support shell commands, Python functions, and remote tools with type annotations. Native MCP support lets Nerve act as both client and server and the README highlights the ability to define MCP servers in YAML to enable agent teams and deeper orchestration. An evaluation mode benchmarks agents using YAML, Parquet, or folder-based test cases with structured logging to support reproducible testing and regression tracking. Workflows let users compose agents into linear pipelines that share context. LLM-agnostic design is achieved via LiteLLM provider support. The project ships as a pip package and has Docker artifacts, documentation, examples, CLI commands, CI, and a GPLv3 license.
Use Cases
Nerve helps developers and teams rapidly prototype and run LLM-driven automation using a consistent, declarative format that improves reproducibility and auditability. The YAML-first approach simplifies defining agent behavior and tooling without embedding code directly in orchestration logic. Native MCP client and server capabilities make it practical to orchestrate multiple agents and build coordinated agent teams. Built-in evaluation tooling enables benchmark-driven development and regression tracking so agent outputs can be validated over time. Workflow composition supports multi-step automations with shared context, and LLM-agnostic model switching reduces vendor lock-in. Packaging as a CLI tool and distribution via PyPI and Docker lowers the barrier to adoption. Documentation, examples, and a community channel are available to support contributors and users.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*