ai-framework

Pocket Flow: 100-line LLM framework. Let Agents build Agents!

The foundation layer for agentic intelligence.

AI-native SaaS framework that builds full-stack apps using autonomous AI agents

A Python-based framework for building multi-agent systems with LLMs. Currently in pre-launch alpha.

Frappe Flow — native AI agents, tools, and triggers for Frappe

CubeLoop — a Pythonic, async-native agent harness core designed for high performance, readability, and production-grade persistence.

PolyCouncil is an open-source multi-model deliberation engine for LM Studio. It runs multiple LLMs in parallel, gathers their answers, scores each response using a shared rubric, and produces a final, consensus-driven result. Designed for testing, comparing, and orchestrating local models with ease.

Framework for building continuous-loop autonomous AI agents.

Turn domain experts into production AI agents.

A Python-Script Based Generative AI platform

Modular multi-agent orchestration framework powered by LangGraph and FastAPI.

A Model Context Protocol server that provides task orchestration capabilities for AI assistants

A Multi-Agentic AI Assistant/Builder

AI Harness & Software Factory

Local-first AI agent framework with GUI, memory, web search, personality constructs, speech i/o, tools, skills, CLI & Telegram features — fully self-hosted via Ollama.

A collection of Summoner clients and agents featuring example implementations and reusable templates

The Mozaiks core is a self-hostable runtime for deploying AI-driven, multi-tenant web apps with real-time UI, auth, and billing built in. Apps are defined via YAML and lightweight Python/JS stubs for agent logic, workflows, and integrations — so developers focus on behavior, not plumbing. 🔗 https://docs.mozaiks.ai/

Medical-AI is a AI framework specifically for Medical Applications https://aibharata.github.io/medicalAI/

JSON Agents - A universal JSON-native standard for describing AI agents, their capabilities, tools, runtimes, and governance in a portable, framework-agnostic format. Based on RFC 8259, JSON Schema 2020-12.

Artifact-first agent framework — disciplined skill chain for AI-driven multi-step engineering tasks. MIT.