# evilsocket/nerve

The Simple Agent Development Kit.

Repository: https://github.com/evilsocket/nerve
Canonical: https://ross.abutalabs.com/products/nerve
Language: Python
License: NOASSERTION
License Family: other
Topics: agent-based-modeling, agents, llm, ai, adk, agent-development-kit, mcp, mcp-client, model-context-protocol
Archived: true
Last push: 2025-08-23T15:12:48+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 38, release rhythm 40, longevity 58
- inputs: {"age_days": 817, "days_push": 375, "days_rel": 400, "gap_med": 1.0, "n_releases_24m": 23}
- flags: archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1316, forks 115 (observed 2026-08-28T04:04:20.819592+00:00)

## What it is
Nerve is a simple Agent Development Kit (ADK) for building, running, evaluating, and orchestrating LLM-based agents using YAML definitions and a CLI. It supports declarative agents, typed tools, native MCP client/server support, workflows, and agent benchmarking, and is LLM-agnostic via LiteLLM.

## Use cases
- build llm agents from yaml files
- orchestrate multi-agent workflows
- run mcp servers defined in yaml
- benchmark and evaluate llm agents
- automate tasks with llm-powered agents
- create multi-step agent pipelines
- connect agents to openai anthropic or ollama models

## When to choose
- you want declarative, auditable, reproducible agent definitions in YAML
- you need MCP client and server support for agent orchestration
- you prefer a CLI-driven, LLM-agnostic agent toolkit
- you want to benchmark and regression-test agents

## When to avoid
- you need a full programmatic SDK in Python rather than YAML-based definitions
- you require complex non-linear agent graphs or advanced orchestration features
- you want a GUI or no-code agent builder

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, mcp, cli, workflow-automation, benchmarking
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: adk, yaml-agents, mcp-client, mcp-server, litellm, declarative-agents, agent-orchestration, agent-evaluation, ai-agents, automation, command-line, docker

## Member repositories
- evilsocket/nerve (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:20.819592+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:48:28.816273+00:00, confidence not recorded.
  - readme: https://github.com/evilsocket/nerve (fetched 2026-08-28T04:04:20.819592+00:00, sha 9209ca29ea0d)
- Data as of 2026-08-30T08:39:29.467469+00:00.