Custom agents

🧠 Curated collection of 214+ best OpenClaw personas — weekly updated by Deskclaw.net from GitHub & community. AI, productivity, dev, marketing, finance & more.

Open-source voice agent OS. Rust runtime, AI-driven builder, sub second latency. Self-host everything.

Shared memory + orchestration for your coding agents — one MCP server, persistent vector memory, agent registry

[ARCHIVED] Superseded by XenolithGraph — modern TypeScript/WebGL successor with React/Vue adapters and MCP for AI agents. See: github.com/XenolithEngine/xenolith-graph

Open-source AI-era employment platform connecting skills, jobs, enterprises, governance, and AI agents.

The AI Agent designed for work with security in mind.

Agent skills for Alpaca's Trading API and Broker API: drop-in SKILL.md files for AI coding assistants

One command in Claude Code, Cursor, or Codex and your agent already knows the physics, the spring curves, and the interaction patterns I've spent years dialing in. Type a minimal prompt and get the interaction you actually wanted instead of something generic.

Local-first AI knowledge system that transforms your accumulated knowledge into an evolving LLM Wiki and knowledge graph. Empower AI agents with your personal knowledge.

MAANG-level AI Engineer interview prep mapped to the AI Engineering Roadmap 2026 newsletter. Deep conceptual, coding, system design & behavioral questions with detailed answers. Covers LLMs, Python for AI, tool calling, agents, RAG, embeddings, vector DBs, eval frameworks & more. Biweekly added with each episode.

An autonomous Site Reliability Engineering (SRE) and DevOps agent for Kubernetes with deterministic log analysis.

Full-stack ecommerce on Cloudflare Workers, D1, and R2 - with admin, Stripe, OpenNode, Lightning payments, and agent-ready APIs.

使用Drl来解决多智能体卸载问题

[ACL 2024] PCA-Bench: Evaluating Multimodal Large Language Models in Perception-Cognition-Action Chain

Deeptoai 系列 RAG 教程

An MCP server that provides safe, read-only access to SQLite databases through Model Context Protocol (MCP). This server is built with the FastMCP framework, which enables LLMs to explore and query SQLite databases with built-in safety features and query validation.

3-tier agentic ChatOps (n8n + GPT-4o + Claude Code) implementing all 21 patterns from "Agentic Design Patterns" — solo operator managing 137 devices

An open-source, code-first Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

A Retrieval-Augmented Generation (RAG) application for querying legal documents. It uses PostgreSQL, Elasticsearch, and LLM to provide summaries and suggestions based on user queries. Features data ingestion with Airflow, real-time monitoring with Grafana, and a Streamlit interface.

PI（π）—— When The Art of War Meets Cognitive Science for Ai

Pydantic AI agent that implements the idea of Claude Skills (progressive disclosure) with no reliance on Claude itself.

Parameterized multi-agent orchestration framework for Claude Code and Gemini

Zero-dependency JavaScript AI chat widget with SSE streaming, search, page context, accessibility, and theming.

基于多智能体协作的中国基金市场智能分析系统与智能管家--A smart analysis system and smart manager for the Chinese fund market based on multi-agent collaboration.

Top papers related to LLM-based agent evaluation

A lightweight, extensible Python framework for building and orchestrating single or multi-agent AI systems. | 一个轻量、可扩展的 Python 框架，用于构建和编排单智能体或多智能体 AI 系统。

A framework for self-improving prediction market trading agents.

Complete Zero-to-Hero Guide for Vibe Coding, AI-Assisted Development, Local LLMs, Frontend, Backend, DevOps, Cloud, Hosting, Debugging, Deployment, and AI-Native Software Engineering.

openJiuwen DeepSearch is a knowledge-enhanced deep search and research framework with industry-leading chunk-level citation and traceable reasoning capabilities, delivering precise agentic search and research abilities.

Persistent memory extension for pi with daily logs, scratchpad, and optional qmd‑powered semantic search.

Edit Videos and Design Images with Claude code or Codex

Claude Code skill: find PR/backlink opportunities on Qwoted, build a sourced statistics page as a linkable asset, and pitch journalists — all from your AI agent.

Local-first agent token usage audit dashboard

The official code implementation of the paper “Soap2Soap: Long Cinematic Video Remaking via Multi-Agent Collaboration“.

An ongoing, collaborative meta-analysis about Human-AI-Interactions. We aggregate data and knowledge to build a non-abrasive, user-friendly prompting framework tailored to LLM mechanics, ensuring reasoning stability and a friction-free prompting environment that is safe for the human psyche and wellbeing.

给智能体安装了思维导图，让 Codex/Claude code/Agent 在复杂问题上用“可见思维树 + 轻量 MCTS”来探索，而不是线性瞎试。

一个基于 LLM Wiki 的 Agent 知识库平台

🐴 Prompt-first harness engineering for safer AI coding agent workflows.

GAN-style self-improvement loop for any text artifact: mutate, grade with a SEPARATE model, keep only verified wins (pairwise-judged), revert the rest. The git history is the improvement log.

AgentKernelArena provides an end-to-end siloed-benchmarking environment where different LLM-powered agents—such as Cursor Agent, Claude Code, Codex, SWE-agent, and GEAK—can be evaluated side-by-side on the same GPU kernel tasks, using objective and reproducible metrics.

A swarm of autonomous AI minds, unified and steered by a single consciousness - the Veil. Runs on any language model.

Desktop automation MCP server — computer use for any AI agent: control screen, windows, mouse/keyboard, and Chrome via Model Context Protocol (stdio)

Your Apple-Native AI Agent

Autopus-ADK is of the agents, by the agents. for the agents. Multi-model orchestration (consensus/pipeline/debate/fastest). Architecture-as-Code, Lore decision tracking, SPEC/EARS engine.

AAAI24(Oral) ProAgent: Building Proactive Cooperative Agents with Large Language Models

A PHP Framework For Creating Autonomous Slack Bots

Chat with PDF lets you ask questions to PDF documents. Built and deployed with NuxtHub, and powered by Cloudflare Workers AI and Vectorize.

RustSBI Specialized Domain Knowledge Quiz LLM

Building a Q&A LLM Agent to Answer Questions about Your Dataset

This is a simple implementation for crypto research agent inspired by Claude Skills repo

An agent-based epidemiology simulation framework built in Rust

AURA (Agent-Usable Resource Assertion) is an open protocol designed to make the web machine-readable. It replaces fragile screen scraping with a declarative aura.json manifest, allowing websites to expose their capabilities as a secure, efficient, and standardized API for AI agents.

This is the source code of "Efficient training techniques for multi-agent reinforcement learning in combatant tasks".

[CVPR 2023] Official repository for FJMP: Factorized Joint Multi-Agent Motion Prediction

Remote MCP server (SEE + Streamable HTTP) implementing the MCP spec's authorization extension. Use directly from your agents, or from Cursor / Claude with mcp-remote

[NeurIPS2021] Federated Reinforcement Learning with Theoretical Guarantees. The repo contains code and experiments for our Federated Policy Gradient with Byzantine Resilience framework for improving sample efficiency of RL agents.

A caching solution for asyncio

An F/OSS solution combining AI with Wikipedia knowledge via a RAG pipeline

Agentic code reviewer for GitHub PRs and GitLab MRs. Multi-step reasoning with autonomous tool orchestration — catches bugs requiring analysis across files.

finance agent + with generative ui

MegaHAL is a learning chatterbot.

Run LLMs and agents on TEEs leveraging NVIDIA GPU TEE and Intel TDX technologies.

A mobile first, self-hosted multi-model AI chat powered by the GitHub Copilot SDK. A modern alternative to ChatGPT, Claude, and Gemini.

Multi-Agent Reinforcement Learning for Mobile Manipulation with NVIDIA Isaac Sim

Astraeus Σ-9000: A Claude Code–native, zero-shot meta-agent orchestration compiler prompt engineered for one-time repository bootstrapping within deterministic, agentic development environments.

A RAG agent using Google's ADK & Vertex AI that lets set up semantic search across documents in under 2 minutes. Features GCS integration and natural language querying

Open-source voice agent orchestration framework - build production voice AI pipelines without vendor lock-in

The NVIDIA NeMo Agent Toolkit UI streamlines interacting with NeMo Agent Toolkit workflows in an easy-to-use web application.

An open-source, headless CRM built for agents. No UI. No dashboard. Just a CLI and a FUSE-mounted filesystem.

One prompt. A full AI engineering team. Go lie on the couch. 🧠

一款轻量级、可扩展的、跨平台的、专为医疗信息化设计的电子病历编辑器内核，为EMR（电子病历系统）提供专业的结构化病历编辑与AI接入解决方案。

This is the official code of the paper "A Multi-Agent System Enables Versatile Information Extraction from the Chemical Literature"

量化投資研究 AI Agent — 透過 CLI 互動介面，自動搜尋財經新聞、分析市場情緒、產生風險評估報告。可接OpenClaw。

Agent Skills for Claude Code and OpenCode

AIRS-Bench: an AI Research Science benchmark for quantifying the end-to-end AI research abilities of LLM agents

TradingAgents-A Share Edition` is a derivative open-source implementation based on [TradingAgents](https://github.com/TauricResearch/TradingAgents?tab=readme-ov-file), focused on China A-share research context and multi-agent collaborative decision workflows.

English translation of MiroFish (github.com/666ghj/MiroFish) - A concise and universal swarm intelligence engine for predicting everything.

人格蒸馏引擎 · 蒸馏自己看清自己，蒸馏亲友留住余温与回声 · Claude Code Skill

MCP server for AI assistants to read and manage Coros fitness data: sleep, HRV, daily metrics, activities, and structured workouts via the unofficial Coros API

Stop your AI from writing like AI. A universal system prompt eliminating every known LLM style tell — works with Claude Code, Gemini CLI, Codex CLI, Copilot, Cursor, and any web AI.

专注于 VLM、VLA、世界模型及通用具身智能等方向，收录前沿论文、开源代码与数据集，追踪从感知到决策的下一代智能体技术。 A curated collection for multimodal intelligence research, covering VLMs, VLAs, World Models, and embodied AI — tracking next-generation agent technologies from perception to decision-making, with a focus on papers, code, and datasets.

Whiteboard tldraw diagrams (.tldr) from natural language with 6 presets and vision-based self-check. PNG/SVG export, multi-agent.

Project Management for the agent era

Animated components, blocks, charts, and templates for React. Copy them in, install with a CLI, or hand them to your coding agent.

Lightweight Codex/Claude loop harness with strict evidence gates.

Build stateful agent workflows with typed outputs, reusable tools, session forks, and ordinary TypeScript.

55 MCP tools to drive the Unity Editor from any MCP-compatible AI agent — scenes, GameObjects, prefabs, assets, tests, and builds. OpenClaw gateway included for remote and chat access.

A股 PEG 估值分析工具 — 彼得·林奇 PEG 投资法本地化实践 | A-share PEG valuation tool with AI-powered analysis

A Claude Code plugin that fills your job applications from your profile, drafts the cover letter, and leaves Submit to you. Locally.

Python web scraping toolkit and MCP server that gives AI agents clean, structured web data from any URL or built-in scrapers.

Gridgeist turns product intent into a rigorous visual system—without the generic AI SaaS aftertaste.

Run multiple Claude Code and Codex sessions in parallel — a browser terminal grid that shows which agent needs you. Local, tmux-backed, MIT.

A repository for skills that are essential to my daily work

Gradient descent, but the parameters are agents — a parallel, asynchronous framework for self-evolving agents (skills, prompts, harnesses). Diffs are the gradients; the aggregator is the optimizer.

[CVPR2024] This is the official implement of MP5

Pilot Protocol: The Internet of Agents

An AI-powered virtual development team of 12 specialized agents that helps you create applications through interactive collaboration. From requirement gathering to deployment, Dream Creator coordinates architects, developers, QA, and DevOps to build your ideas into reality. Supports Chinese and English languages.

The goal of this tool is to help you to have fast responses when you have some problems with OVH services.

Google's Meena transformer chatbot implementation