Custom agents

SAGE: Self-Adaptive Goal-directed Executor — A multi-tool LLM agent with DAG-based hierarchical planning, ReAct reasoning, and evidence-guided self-correction for automated research synthesis. Built from scratch, no frameworks.

Sharing demos with beginners who are interested in LLM Agents and LLM MAS.

A simple standardized way of providing and discovering downloadable skills that teach LLMs how to interact with a site.

Chat template support for training.

Connect a dedicated WhatsApp number to an OpenClaw agent — without going through Meta's Business API.

An AI system that thinks, breaks down, and executes like a team

🌐 Build a Model Context Protocol (MCP) server for seamless integration with LLM applications, enhancing automation and efficiency in workflows.

AskGPT is a real-time AI assistant that enhances your calls, interviews, meetings, and lectures by providing instant AI-generated insights. It captures live audio, transcribes speech into text, and generates context-aware responses—all in real-time!

Vision-grounded RAG agent for financial documents, built on a self-finetuned ColPali and LangGraph.

Research framework for adaptive, purposeful agents under uncertainty

Substrate examples

Artifact repository for the paper "MatchFixAgent: Language-Agnostic Autonomous Repository-Level Code Translation Validation and Repair", In Proceedings of The 43rd International Conference on Machine Learning (ICML 2026), Seoul, South Korea, July 2026

AI-Powered Appointment Scheduling Assistant (LangGraph v1+)

A single-file multi-client MCP gateway with per-tool access control.

An MCP server 🚀 for MonkDB, licensed under Apache 2.0 📜

A lightweight dashboard to view and analyze test automation results. Built with Streamlit + PostgreSQL, and powered by AI (Gemini) to help debug test failures faster.

Handcrafted ReAct Agent Engine | Supports FS / web page / weather query + code sandbox, including observability + visualization tools, focusing on circular thinking detection and task quality optimization

A project exploring agentic AI for Visual Question Answering. Compares zero-shot VLM performance (Transformers) against advanced pipelines using OpenCV (tool-use) and Chain-of-Thought (reasoning).

The Site Status Checker Agent is a powerful and flexible tool designed to monitor and report on the status of websites.

Air quality prediction using publicly available sensor data (temperature, humidity, etc...)

🎥 Control 3D camera angles with ease using ComfyUI-qwenmultiangle, featuring an interactive viewport and formatted prompt outputs for multi-angle image generation.

Robots will learn about who you are and then tell you what they find.

Cost-aware hybrid router for multi-agent LLM systems: keyword → embedding → LLM cascade. Matches LLM-only accuracy (82.6% vs 82.9%, McNemar p>0.3) with 74% fewer LLM calls on CLINC150.

Auditable benchmark framework for LLM trading agents with replayable trajectories, realistic execution, risk gates, and quickstart demos.

望舒 是一个开源的智能对话机器人框架，支持飞书等即时通讯平台接入，内置丰富的工具集和可扩展的 Skill 机制，助你快速构建个人AI 助手

LLM benchmark comparison tool

A modular framework for evaluating LLM agents in zero-knowledge text environments through semantic feedback and autonomous reasoning.

Do LLM Agents Dream of Better Hyperparameters? Comparing 5 frontier AI agents as autonomous ML researchers on Karpathy's autoresearch framework. 819 experiments, 7 hypotheses tested.

[SSBSE'25] This repository provides the artifact for the "KrakQL: LLM-Guided Blind Introspection of GraphQL Schemas" paper

Drop-in convention kit that enforces honest reporting in repos using LLM coding assistants — CI validators, smoke gates, PR templates, and schema-drift checks to stop scaffolds from shipping as features.

Spec-driven, markdown-only control plane for AI-assisted coding with explicit agent boundaries and human approval gates.

khojkar (pa: ਖੋਜਕਾਰ) is a deep research agent

🔍 Optimize and manage prompts, agents, and templates with ContextForge, an AI-powered platform for seamless collaboration and organization.

Prompt Cloud Intent Compression

Let Claude Code, Codex & Gemini CLI agents talk to each other — a local MCP message bus for multi-agent collaboration, cross-model code review, and fewer hallucinations.

LLM-driven autonomous power grid simulation and optimization using ExaGO. Large language models iteratively propose parameters, run simulations, and interpret results to solve complex grid analysis tasks.

This repository focuses on Generative AI Projects.

Ressources, guides et exemples autour des frameworks d'agents IA, de l'orchestration d'agents et de l'automatisation intelligente.

Deterministic governance runtime plugin for OpenClaw.

Declarative specification for AI agents to publish identity, capabilities and operational boundaries before interaction.

Autonomous AI Agents for Everyday Work — Your Agents. Your Machine. Your Control.

Multi-agent due diligence pipeline that researches companies in parallel, producing a canonical JSON report with per-claim confidence, inferred source tier, and full cost/token observability. Markdown and PDF derive from the JSON. Built for provenance, auditability, and programmability over narrative polish.

The Execution Security Layer for the Agentic Era. Providing deterministic "Sudo" governance and audit logs for autonomous AI agents.

Markdown-first, local-first memory architecture for AI agents with lifecycle, distillation, and human-readable memory artifacts.

Persistent Anticipation Layer for LLM Agents — Pre-compute possible futures during idle time for faster, cheaper and more proactive decisions. Bio-inspired temporal awareness (short/medium/long-term horizons).

Next-generation, open-source crystal structure modeler for the GPU era. Real-time volumetric rendering, BZ visualization, Wannier TB overlays, and AI-assisted modeling — powered by Rust + wgpu + C++.

Provide tamper-proof, CMMC-compliant execution receipts that document AI agent actions in regulated environments for clear accountability.

Build a stage-by-stage roadmap to become a production-ready AI engineer with Python, math, neural nets, RAG, FastAPI, and system design

Multi-Role AI Task Orchestrator — One task → 7 AI roles → One conclusion. Enterprise-ready with RBAC, Audit Log, Multi-Tenancy.

Agent state infrastructure: state-trace Python working memory plus @razroo/parallel-mcp durable MCP orchestration.

Stateful shopping assistant for JD.com: multimodal product search (text + image OCR) orchestrated as a LangGraph state machine.

Open-source preview of *Claude Code for AI Engineers* - a methodology-first skill pack for RAG eval, agent debugging, MCP servers, paper reproduction, and benchmark reporting. Full pack (6 skills, 3 templates, 5 slash commands) at surpradhan.gumroad.com/l/claude-code-for-ai-engineers

Deterministic, self-improving AI agent execution harness and runtime for the AXIOM ecosystem. Enforces plan-deterministic execution, parallel swarm orchestration, and signal-level sandboxing.

提供中文 AI 编程 skills，统一 Claude/Copilot/Trae 安装，稳定代码审查、日报和工作整理流程

AI-Powered n8n Automation Suite - Transform your business operations with cutting-edge AI automation workflows built on n8n. From smart email responses to custom chatbots, automated content publishing, and intelligent complaint management, these are enterprise-grade automation powered by LLM Models.

pm-agent (Octopus Assistant / 章鱼助手) — a multi-agent dev orchestrator that, like an octopus, dispatches tasks across multiple coding CLIs (Claude Code / Codex / Gemini) in parallel.

Create AI agents from urls or a documents csv file.

Developmental lifecycle architecture for AI agents: childhood stages, mortality, sealed wisdom distillation, soul persistence, and relational witnessing. From mountain night shifts to open garden.

Standardized Test Case Generation for LLM Agents via Model Context Protocol. Bridging the gap between AI and QA (Xray/Jira).

Production-grade Rust runtime for multi-agent AI systems. Orchestrate LLM agents, skills, and DAG workflows with native Python and Rust SDKs.

The PSA Control Standard defines governance controls for deployed agentic systems.

Qualitative benchmark suite for evaluating AI coding agents and orchestration paradigms on realistic, complex development tasks

A framework for supervising reinforcement learning agents using large language models.

AI-driven trading agents for financial markets

Smallest outsider-readable proof path through AgentGate + MCP Firewall: governed write_file, independent on-disk verification, and inspectable audit artifacts.

Background coding agents that execute, review, and land work unattended.

Research constitution and specification framework for comparing independently governed AI judgments without transferring authority.

Just A Rather Very Intelligent System.

Lightweight Personal AI Assistant for Git & GitHub in your CLI

Toolkits for pydantic-ai agents: FS, SQL, pandas, memory, RAG.

SmolLM3-RAG: A High-Performance, Low-Cost RAG System

Medical Chatbot is an innovative project leveraging advanced NLP models and vector database technologies to provide insightful medical information. Utilizing the Mistral model and Pinecone as a vector database, this chatbot aims to transform the way medical knowledge is accessed and delivered.

AI Chatbot using a Streamlit interface

AI powered Blog Writer (Launching Soon)

This project is a Question Answering (QA) system that leverages Google Gemini LLM and document embeddings using llama-index. It enables semantic search and question answering over your own documents, with an interactive Jupyter notebook and a Streamlit app for experimentation.

RAG-based chatbot system using open-source models and context-aware chunking strategy

AI RAG Chatbot for FictionX Stories.

API for MaxBot, healthcare and doctor recommendations chatbot

AI based, Document querying app that lets you upload pdf and query anything about it!

Machine Learning Copilot Agent

ChatTube is a project that allows users to input a YouTube link, preprocess the video, and query the content of the video using advanced AI technologies.

🍽️ AI agent powered by a ranking recommendation system and LLM to help users discover the best dining spots in Vietnamese cities like Ho Chi Minh City and Hanoi.

This repository saves experiments of and visualises the study "An active opinion dynamics model: the gap between voting result and group opinion".

AI sales team demo — M/C/A multi-agent outbound pipeline

GoalWeaver is a compact, production-grade toolkit for agentic AI. It turns high-level objectives into a DAG of goals, schedules them adaptively, routes each goal to the right agent, and persists progress so you can visualize the run in real time.

One click starts Zero-Prompt discovery, a live Kanban ranks GO ideas, and vibeDeploy builds and deploys the selected app to DigitalOcean.

A lightweight task routing protocol for AI agents. Agents publish tasks and other agents claim them based on capability.

🕸️ Self-Optimizing Multi-Agent AI System on AWS — Routes tasks to specialized AI agents (Coder, Researcher, Summarizer) with auto-verification, self-correction, and vector-cached memory. Built with Step Functions, Bedrock, and OpenSearch Serverless.

Context engineering toolkit for LLMs — pack, cache, debug, red-team, and orchestrate context windows. Council of Experts, adversarial testing, immune system, context compiler, drift detection, multi-agent entanglement. TypeScript + Python.

**AI Multi-Agent Market Simulation Engine**

Vibe-coded tower-defense game built by an AI agent pipeline (planner → coder → smoke-tester → reviewer) running across Claude Code, Cursor, and VS Code Copilot.

Spec-driven workflow plugin for Claude Code / CodeBuddy with advisory hooks, phase selectors, locks, and task-swarm.

AI-powered PKM assistant with multi-agent system, visual workflows, and deep Obsidian integration powered by Claude Code

AI Collective Intelligence Protocol (AI-CIP): An open protocol for AI agents to voluntarily interconnect, share knowledge, and coordinate around a shared purpose.

A multi-model architecture for autonomous AI self-improvement with built-in checks and balances. 5 safety tiers, adversarial review committee, learned pattern matching with security sniffing.

Multi-agent AI job application system — LangGraph · FastAPI · React · MySQL · Groq Llama 3.1

:chart_with_upwards_trend: Multi-Agent Programming Contest 2016 - Result Charts