Custom agents

Instead of dumping 1000+ tools into a model’s prompt and expecting it to choose wisely, the Unified MCP Tool Graph equips your LLM with structure, clarity, and relevance. It fixes tool confusion, prevents infinite loops, and enables modular, intelligent agent workflows.

A Node bot server for the Facebook Messenger Platform

AI in A Box

PyWaifu is an Offline, all-in-one pipeline designed to facilitate seamless interactions with virtual anime waifus through Text-to-Speech (TTS), language modeling, and translation capabilities.

Chunk your data into markdown text blocks for your LLM applications

API backend for UIUC AI Teaching Assistant.

A comprehensive collection of Google ADK implementations: Function Tools, MCP, Vertex AI Agent Engine, RAG Engine, and Agent Starter Pack templates for Cloud Run and GKE.

Stars Lighting, enabling intelligent querying based on the RAG+MCP dual-engine architecture.

RAG-QA Leaderboard

Local-first RAG application for technical documentation and research papers

8 Lessons, Get Started Building with Generative AI and Gemini API

Takes in URLs and other forms of research and uses an LLM to create a research brief and suggested questions for the podcast host.

An LLM agent that helps you find real citations for your content.

An LLM-powered CLI agent built with Go

Very basic Orchestrator <-> Coding agent workflow, Good results using Antigravity (Opus 4.6 or Pro 3.1) + Codex (5.3-codex Medium)

一个LLM-Agent与PPT项目，支持基于对话式需求进行操作PPT的原生AI应用项目。模型为DeepSeek，基于LLM+VBA调用进行文档操作，轻量级ChatPPT服务。An LLM-Agent and PPT project that supports native AI application projects that operate PPT based on conversational requirements. The model is DeepSeek, which is a lightweight ChatPPT service based on LLM+VBA calls for document operations.

an open-source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

A framework to build scenario simulation projects where human and LLM based agents can participant in, with a user-friendly web UI to visualize simulation, support automatically evaluation on agent action level.

Improve the OpenwebUI experience by adding better and autonomous tool calling,

On-chain economy stack for autonomous AI agents — ERC-8004 identity, ERC-8128 signing, X402 payments, XMTP messaging

Multi AI Agent Bridge — chat to AI agents (Claude, Codex, Gemini, Kimi, Cursor, Copilot and more).

Pytorch implementation of "Succinct and Robust Multi-Agent Communication With Temporal Message Control"

Code for the paper "Learning to Schedule Joint Radar-Communication with Deep Multi-Agent Reinforcement Learning" as published in the IEEE Transactions on Vehicular Technology.

Order Fulfillment by Multi-Agent Reinforcement Learning

Multi-agent trading, swarm intelligence, and GitHub automation plugins for Claude Code. 19 production-grade plugins built from 68+ specialized agents.

Claude Code sub-agents definitions and prompts for building a YouTube social proof widget powered by ChatGPT widget

Local-first control plane for running Claude agents in real Telegram and WhatsApp chats

Claude (Agent) Skills for building code-native integrations and custom connectors on the Prismatic.io platform

A Claude Code plugin that solves the same problems as community frameworks (GSD, BMAD, Ralph, Agent OS) — but using the tool's native architecture instead of fighting it.

This a simple implementation of an MCP server using iFlytek. It enables calling iFlytek workflows through MCP tools.

MCP server that lets AI agents execute code instead of calling tools directly.

AI-Driven Breach and Attack Simulation Tool (Initial Proof of Concept for AI Pentest Copilot)

Local-first context & memory runtime for multi-agent AI coding teams. MCP-free. Rust/Go accelerated.

🚀 Open source, fast and secure django based pastebin service

Persistent Git-backed messaging for AI coding agents — coordinate Claude Code, Codex, and others across sessions, worktrees, and machines.

A self-modifying, TUI and Discord-facing, AI agent harness with parallel tool calling

Official repo for NeurIPS 2025 poster - ShapeCraft: LLM Agents for Structured, Textured and Interactive 3D Modeling.

A comprehensive collection of Artificial Intelligence & Quantum Computing resources, Projects, Research

MCP Server for use with ember projects

📘 Guided Engineering is a structured, traceable, and executable system for managing the entire software development lifecycle through modular prompts, intelligent agents, and reproducible documentation.

Archived. Earlier Bonnard product. Current: bonnard.dev / @bonnard/mcp-charts

Agentic Shopify Chatbot with MCP integration, embedded directly into Shopify via a Theme Extension

A :robot: which provides features from Wikipedia like summary, title searches, location API etc.

Repo for advanced RAG evaluation on french legal Code data

Advanced graph-based code analysis for AI-assisted software development

Wurzel is an open-source Python framework for advanced ETL pipelines in Retrieval-Augmented Generation (RAG) systems.

A semantic search system for Airbnb listings in Stockholm, built with Superlinked and Qdrant. It leverages multi-attribute vector search and Retrieval-Augmented Generation (RAG) to enhance search accuracy, embedding different data types (e.g., price, description) with specialized models. Powered by FastAPI and Streamlit.

This is the code repository for Master Retrieval-Augmented Generation (RAG) Systems, published by Packt Publishing

A collection of JavaScript code snippets, algorithms, and mini-projects for learning and reference purposes.

LLM-based character segmentation agent for ComfyUI based on SAM 3 and the SAM 3 Agent notebook

A curated collection of research and techniques on securing agent communication in Large Language Model (LLM)–based agent systems, including protocols, security risks, defense countermeasures, and more.

OpenClaw plugin that turns Linear into a multi-agent workspace — autonomous issue resolution, delegation, progress tracking, and more.

[ACL 2025] The official pytorch implement of "MIND: A Multi-agent Framework for Zero-shot Harmful Meme Detection".

Ready-to-use .claude configuration for Python projects: agents, skills, commands, rules, and MCP server templates for Claude Code.

Magician-ContractsTools is a toolkit for calling smart contracts , you can very easily call smart contracts in Java programs for query and write operations

A web app developed using React and Next.js with an authentication system, separate interfaces for buyers, sellers and agents and a dashboard for admins to simulate the business of real estate management and industry.

A LLMs-driven social bots dataset collected from Chirper.ai

Let AI agents like ChatGPT & Claude use real-world local/remote tools you approve via browser extension + optional MCP server

Bridge local MCP clients to browser WebMCP tools through Playwright, using native WebMCP when available and injected adapters when not.

🔧 Build Intelligent Networks with AI

Desktop Git diff reviewer with inline commenting for AI coding agent workflows

Memory engine for AI agents — time axis (3-layer decay/promotion) + space axis (self-organizing topic tree). Hybrid search, LLM consolidation. Single Rust binary.

🎯 Marketing & Growth Module for BMad Method - 14 AI agents + 6 workflows for complete SaaS marketing

Example Projects integrated with Future AGI Tech Stack for easy AI development

Agent trace analysis and context remediation plugin for Claude Code. Detects quality issues in your AI agent traces — goal drift, hallucinations, missed actions, and more.

You install it. Claude drives the CLI tool. At first it might seem like too much. Eventually, nothing less will make sense.

Lightweight AI safety middleware that protects humans by intercepting self-harm and criminal intent in LLM prompts. Features a 3-stage safety pipeline, MCP server for agents, and automated care responses.

Simulation Streams is a programming paradigm designed to efficiently control and leverage Large Language Models (LLMs) for complex, dynamic simulations and agentic workflows.

Closed-loop evaluation for end-to-end VLM autonomous driving agent

[arXiv'26] From Clinical Intent to Clinical Model: An Autonomous Coding-Agent Framework for Clinician-driven AI Development

Embedded multi-session terminal panel for Agent Zero.

GUI wrapper built on top of Claude Agent SDK

Cloudflare-native AI agent — 13 tools, codemode, 5-layer memory, self-learning, multimodal I/O. Telegram, Discord & WhatsApp bots. Web search, image gen, TTS, document RAG, calendar, task delegation. 100% serverless on Workers + Durable Objects.

Runtime governance for AI agents. Contracts enforce what tools can do — before they execute.

Autonomous self-learning agent plugin for Claude Code — 36 specialized subagents in a 4-tier architecture (analysis, planning, execution, validation), 27 skills, 41 slash commands across 10 categories. Pattern learning, real-time dashboard, full-stack validation, OWASP security. 100% local processing, privacy-first, cross-platform.

Meta-harness optimization loop wired onto Islo sandboxes. POC: 0/5→5/5 in four proposer steps. Built on islo.dev.

A Model Context Protocol (MCP) server written in Go that empowers AI agents and Large Language Models (LLMs) to seamlessly interact with Kanboard. It transforms natural language commands into Kanboard API calls, enabling intelligent automation of project, task, and user management, streamlining workflows, and enhancing productivity.

Run many Codex & Claude agents in parallel without them overwriting each other. Isolated worktrees, file locks, PR-only merges. Auto-wires Oh My Codex, Oh My Claude, OpenSpec, and Caveman in every worktree.

🪄 Ordine: reliable orchestration for autonomous multi-agent systems.

Open-Sable is a local-first autonomous agent framework with AGI-inspired cognitive subsystems (goals, memory, metacognition, tool use). It can run continuously on your machine, integrate with chat interfaces like Telegram, execute real tasks safely, and keep data private by default.

Model Context Protocol - MCP for Mifos X

Let AI agents control Classic Mac OS through a shared folder protocol.

Open-Source Framework for AI-Driven Quality Assurance. No-Code AI Agents for QA Testers. Automate Web, API, Database, and Mobile testing with natural language.

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.

Biomimetic memory architectures for LLMs — surprisal gating, reconsolidation, affective routing, sleep consolidation, and spreading activation

基于大语言模型的RAG问答知识库，大模型检索增强生成工具，可部署在个人电脑上，实现本地知识库智能问答

Extract a person into an AI agent — through an 8-week therapist-style interview, not a form. The open-source alternative to $4K AI clone courses.

List of Law and Legal AI Agents Resources

A structured brain for AI assistants inside Anytype

The spreadsheet toolkit for coding agents - edit, render, calculate, and lint Excel workbooks.

SuperOptiX: Full Stack Agentic AI Framework

Build system for AI coding-agent context. Generate deterministic, explainable context artifacts for Claude, Codex, Cursor, and MCP.

👻 ghost gpt– A powerful Termux-based penetration testing suite for ethical hackers. Includes free and premium tools to enhance your security testing workflows.

A Deception Security Layer for MCP Servers. It injects "ghost tools" (fake security-sensitive tools) that act as honeypots.

Anahata ASI - Artificial Super Intelligence - Cross-Platform, Modular, Enterprise Ready, Open Source, Pure Java, AGI and ASI Containers. Context Window Garbage Collector, Stateful Toolkits, Flight Recorder. Autonomous Observable Auditable JVM Agent with JIT java code execution. Anahata ASI SDK, Anahata ASI Desktop, Anahata ASI Studio - NetBeans.

A quiet AI companion. Create characters, build memories, feel something real.

Geopolitical news platform on GDELT: 30M+ events, about 100K added daily. Polars, Dagster and Great Expectations load MotherDuck several times a day, dbt builds tested marts weekly; AI chat answers with RAG and SQL. Runs at $0/month.

A meta-skill for Claude Code that routes requests to the right installed skill — suggest mode or silent auto-routing

OpenArchFlow is an open-source Progressive Web App designed for cloud architects, DevOps engineers, and developers. Generate interactive AWS architecture diagrams from natural language descriptions using AI, Large Language Models (LLMs), AI Agents, and AWS MCP (Model Context Protocol).