praesidia-ai

praesidia-ai / neurogent

Public

The open-source framework and CLI for building, packaging, and deploying A2A-compliant AI agents — with built-in Praesidia security.

13
2
100% credibility
Found Mar 18, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Neurogent is a multi-agent terminal shell and open agent framework that enables users to run teams of AI specialists in the terminal using YAML configurations and interact with them via commands, supporting models like Anthropic or OpenAI.

Star Growth

See how this repo grew from 13 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is neurogent?

Neurogent is an open-source AI agents framework in TypeScript that lets you build, deploy, and run teams of LLM-powered agents compliant with the A2A protocol for seamless discovery and communication. Install via npm, scaffold projects with `neurogent init`, define agents in YAML with tools and models (Anthropic, OpenAI, Ollama), then dev locally, build AgentCards, and deploy Docker containers. It also includes a terminal shell (`neurogent-shell`) for chatting with pre-built agent packs like dev-trio or full-team, supporting mentions (@agent), chaining (>>), and swarm mode.

Why is it gaining traction?

Unlike LangChain or CrewAI, which focus on orchestration logic in code, Neurogent emphasizes runtime and deployment: YAML configs swap models without rewrites, one-command Docker deploys, and native A2A endpoints for agent-to-agent calls. The terminal shell delivers instant multi-agent collaboration—no backend setup—while Praesidia hooks add policy-based security like jailbreak blocking and PII redaction out of the box. Developers dig the self-hosted, production-ready flow for AI workflows.

Who should use this?

Backend engineers building secure, composable AI tools for code review chains or research pipelines. DevOps teams deploying self-hosted LLM agents as microservices. Startups prototyping agent teams for engineering, marketing, or support without vendor lock-in.

Verdict

Early alpha with 13 stars and 1.0% credibility score—test coverage and stability need work—but the README and CLI make it dead simple to try A2A agents today. Worth forking if you're into open-source AI frameworks; skip for mission-critical prod until more battle-tested.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.