Custom agents

Pretrain、Posttrain、RAG、Agent等大模型相关的基础项目合集

This Coding Agent is a multi-LLM engineering crew that plans tasks, writes code, and self-reviews to build polished outputs like the ArXiv CS Daily site—featuring automated paper ingestion, intelligent tagging, UI generation, and full audit logs in one reproducible workflow.

SimpleLLMs - Simple LLM Suite: Five specialized agents for Claude Code (R.A.L.P.H., B.A.R.T., L.I.S.A., M.A.R.G.E., H.O.M.E.R.)

Deep learning models for self-driving vehicles to predict other car/cyclist/pedestrian (called "agent")'s motion.

Adaptive Traffic Signal Control of Autonomous Vehicles using Multi-Agent Reinforcement Learning techniques and SUMO tool for vehicular simulation. Implemented algorithm: Independent Advantageous Actor-Critic (IA2C)

Multi-agent reinforcement learning environment

多智能体系统一致性

A flexible framework for multi-agent models in economics and finance

Implementation of Multi-Agent Deep Deterministic Policy Gradients

MultiROS is an open-source ROS based simulation environment designed for concurrent deep reinforcement learning. It provides a flexible and scalable framework for training and evaluating reinforcement learning agents for complex robotic tasks.

Multi-Agent Reinforcement Learning for Traffic Signal Control through Universal Communication Method

Implementation of "Federated Control with Hierarchical Multi-Agent Deep Reinforcement Learning" (https://arxiv.org/pdf/1712.08266.pdf)

CTRMs: Learning to Construct Cooperative Timed Roadmaps for Multi-agent Path Planning in Continuous Spaces (AAMAS-22)

This is the official implementation of "Optimizing Large-Scale Fleet Management on a Road Network using Multi-Agent Deep Reinforcement Learning with Graph Neural Network" (ITSC 2021)

[AAAI 2023 Oral] Contrastive Identity-Aware Learning for Multi-Agent Value Decomposition

We explore the application of deep reinforcement learning in the field of robotic control, the cooperative and competitive behavior of multi-agents in different game types, including RPG and MOBA, cloud infrastructure, and software engineering as well.

A specialized Claude Code agent that addresses Claude's overly lenient validation standards and visual hallucinations when assessing UI modifications. Provides strict, skeptical secondary validation to ensure accurate visual verification of code changes.

Evolutionary algorithm discovery using Claude Code

An existential western RPG about semi-conscious GPT agents. Written in a 3d engine from scratch targeting Wasm/Webgl and Desktop

A WIP framework for developing cross-platform chatbots

Reinforcement Learning using the Actor-Critic framework for the L2RPN challenge (https://l2rpn.chalearn.org/ & https://competitions.codalab.org/competitions/22845#learn_the_details-overview). The agent trained using this code was one of the winners of the challenge. The code runs on the pypownet environment (https://github.com/MarvinLer/pypownet). It is released under a license of LGPLv3

An agent-based modeling toolkit written in Python.

Agent Memory Playground: AI Agent Memory Design & Optimization Techniques

Bridge WhatsApp with Claude Code - interact with your files via WhatsApp messages

Context Strategy Framework: Intent → Build → Learn workflow. Preserves context in AI-generated code.

Memory management system to enhance AI agents with personalized, reasoned and context-aware responses

Markdown-first work-memory protocol for existing agents, with maintained knowledge, candidate notes, evals, and an example KB.

Convert HTML to Markdown with Elixir

Hierarchical RAG architecture scaling to 693K chunks on consumer hardware (4GB VRAM). Features 3-address routing, hybrid vector+graph fusion, and SetFit classification.

Drop-in pytest plugin for regression-testing AI agents — snapshot baselines, semantic comparison, mock LLMs

AI Agent - It can book rides, order food, post tweets and also control basic tasks on device.

For testing agent capabilities

Collection of n8n workflow automation JSONs for various use cases, including AI agents, integrations, and automated processes. Ready-to-import workflows with complete documentation.

Solving Computer Vision with AI agents

Neobanker FinTalk-AI: A Grounded Orchestration Framework for Multi-Agent Collaboration on Financial Tasks Leveraging the OSWorld Environment

An RL Recipe for Building Agentic LLMs via Self-Imitation on Long-Horizon Agentic Tasks

Minimal autonomous coding agent MVP — drop user stories in inbox/, get git commits out. Built on top of pi-coding-agent.

异构混合阶多智能体系统编队控制的分布式优化

Pi skill for multi-agent workflow orchestration with file-based handoff

MATLAB code for the numerical example in ``Event-Triggered Consensus for Multi-Agent Systems with Guaranteed Robustly Positive Minimum Inter-Event Times

A barebones multi-agent orchestration tool for claude code.

a novel framework for stock portfolio trading that employs a 'Relaxation and Refinement' strategy to boost the Soft Actor-Critic (SAC) agent

Universal MCP server for AI agents to interact with any* blockchain via natural language and plugins. Supports swaps, bridges, gas, staking, lending, and more across Ethereum, Arbitrum, Base, Polygon, BSC, and testnets.

Stop typing. Start autofilling with Superfill AI

OpenClaw plugin for persistent AI memory

Open source product copilot. Embed an AI chat widget that doesn't just answer questions—it executes client-side actions in the user's browser to get things done inside your app.

AI chatbot used to communication with End user through online on platforms such websites and application. Btech college project. Project Includes Source Code, PPT, Synopsis, Report, Documents, Base Research Paper & Video tutorials

World Best Multi Language Whatsapp Advanced Chat Bot

🔥 Alternative to Ollama — multi-model serving with sub-ms model switching · CPU-only 20B inference for Edge AI · llama.cpp + stablediffusion.cpp

AI Multi-agent system for real-time, adaptive supply chain coordination and optimization leveraging responsive AI clusters.

Markdown files for managing a coding AI Agent

A curated list of emerging conventions to build AI agents.

User profile and long-term memory for your AI agent. Drop-in LLM proxy that gives AI model persistent memory and structured user understanding.

An AI-powered solution designed to enhance mobility for visually impaired individuals by combining real-time video analysis with voice-guided navigation. Also features a built-in Multi-Agent AI Assistant — users can speak naturally to complete map navigation, adjust system settings, send messages to family.

An iOS SDK for building LLM agents.

VideoMultiAgents: A Multi-Agent Framework for Video Question Answering

A system for running Multi-Agent Path Finding (MAPF) experiments, with multiple implemented algorithms.

This is a python version of Claude Code

A brainstorming framework for Agentic harnesses (e.g. Claude Code, Opencode, Copilot) that challenges you instead of agreeing with you.

Eval Protocol (EP) is an open solution for doing reinforcement learning fine-tuning on existing agents — across any language, container, or framework.

Context-Driven Development for GitHub Copilot

Universal skill installer for AI coding agents

A complete AI-powered organizational structure: 42 agents, 10 playbooks, one company.

Refine UI directly in the browser and feed the changes to your coding agent.

An AI-powered autonomous reverse engineering swarm that accelerates binary analysis through multi-agent collaboration

Home Assistant Conversation agent using xAI Grok LLM

Agent-readable second brain using PostgreSQL + pgvector for semantic search and memory storage.

AI system powered by large language models.

Vapi CLI - Voice AI agents for developers

Agent runtime with full system prompt control. Define agents in markdown, run them in a terminal TUI or multi-user web UI. Powered by the Claude Agent SDK.

Open-source, self-hosted AI agent that evolves its own code to perform better in your specific use-cases.

A multi AI Agents program that generates trading signals

A coordinated team of 53 Claude Code subagents that build, ship, and operate mobile apps end to end. Each agent owns a narrow slice of the lifecycle so that responsibility, expertise, and review checkpoints stay separated and observable.

AI Agent Skills toolkit for automated product introduction video generation with Remotion, Playwright, and edge-tts

Memory layer for AI Agents and OpenClaw powered by Rust

这是一个为 AstrBot 设计的 Office 助手插件。它赋予大语言模型（LLM）直接操作文件的能力，支持读取并分析多种格式文件，以及生成 Office 文档和office互转pdf的功能

Built for OpenClaw and beyond, Chat Selfie lets your AI Agent send you emotional selfies in every conversation, with mood-aware replies, heartbeat pushes, and flexible image generation. 让你的 AI Agent 在对话中发送情感自拍。

CLI-based multi-agents for Auto-Tuning (e.g. HPC code optimazation loops) supporting Local LLMs

CLI for Monarch Money - access your financial data from the command line. Built for agents(claude code) and humans.

Honeydew AI plugins for coding agents (Claude Code, Codex, Cursor, Github Copilot, Gemini, etc...)

query search engine cli for humans and ai agents

Transforme Claude Code en générateur autonome de knowledge vault Obsidian. Pattern Karpathy. Par LE LABO IA.

A full-stack conversational AI starter kit built with Amazon Bedrock AgentCore, Strands Agents SDK, FastAPI, and htmx.

RAG pipeline security testing toolkit - 27 techniques across 6 kill chain phases, mapped to MITRE ATLAS

Spec-driven development (SDD) plugin for Claude Code — a collection of specialized AI agents, phased implementation plans, and verified code generation for full-stack teams

A demonstration of hybrid search with reranking using Qdrant and BGE-M3 model. A showcase of dense and sparse retrieval combined with ColBERT reranking for optimal search results

Kumo Coding Agent — context docs, skills, and self-maintenance for building ML models with the Kumo SDK

Use Claude Code Agent Teams inside Zellij — a tmux shim that translates tmux commands to zellij actions

Open-source platform connecting AI assistants to government open data — MCP server, curated civic MCP directory, and anti-hallucination framework for all 559 Socrata portals

Claude Skills For Sales Optimization

A schema standard for LLM-compiled personal knowledge bases. AGENTS.md spec, templates, worked example, spaced repetition learning layer.

Minimalistic implementation for LLM-based chat assistants with Tool Use (function calling) and MCP

40 AI agent skills for VTEX platform development. Install with: gh skill install vtex/skills. Exports to Cursor, Copilot, Claude, AGENTS.md, OpenCode, and Kiro.

Entire suite of VS Code Copilot agents, skills, instructions and prompts for creating modern wow addons

Put Codex in your WhatsApp with QR auth, MCP tools, and optional phone-based control.

Web search CLI for AI agents — 13 providers rank-fused into one Rust binary. Search, extract, verify, check credits.

A Local add-on that provides an MCP server, skills, and project context for AI-powered WordPress development. Works with Claude Code, Cursor, Windsurf, VS Code Copilot, and any MCP client.