Custom agents

Contract-based multi-agent skill framework for Claude Code & Codex — compile YAML runbooks into SKILL.md with loop, parallel, and checkpoint support.

Meta·MFH·Orchestra proof-bounded AI OS — 실행 주장은 evidence gate를 통과한다.

AI-powered Web3 chat that mixes live market and DeFi data (CoinGecko, DeFiLlama) with RAG over a PostgreSQL/pgvector knowledge base; React + Vite frontend and Express backend.

Read-only AI trading companion harness for decision logging, D1/D3 review, and rule evolution.

100% local research agent for your papers - ask PDFs on your machine, page-accurate verified citations, cross-paper synthesis, Ollama-ready

Ai-LaTex: A local-first LaTeX editor with PDF compilation, AI Agent collaboration, Codex/Claude Code support, and fine-grained academic writing assistance.

F1 Race Report is an automated tool for generating insightful email race summaries the morning after a Grand Prix weekend has taken place! 🏎️

Local task-state ledger for AI coding agents. Preserve decisions, source conclusions, evidence, and checkpoints across Codex, Claude Code, Cursor, and other MCP clients.

Who is Naboraj Sarkar? 🚀 Student, creator & tech learner from India 🇮🇳 — projects, journey, skills, and digital presence in one place.

A structured framework for giving AI assisted teams the architectural context they need to generate code that fits your system.

Agents that learn by doing — and remember how they did it. A methodology for AI agents to create, evolve, and share reusable skills.

Cactus AI framework plugin for UE5 to run local LLMs at runtime. https://github.com/cactus-compute/cactus

A helper tool to sync a tree of Jira tasks into beads tasks, and sync beads state back to Jira

Actor-based DAG workflow execution engine for low-code applications and Agentic AI

Delegate hard problems to a more capable advisor model while working with a cheaper, faster model day-to-day

Self-hosted Dakera AI memory server — Docker Compose, Kubernetes, Helm, HA cluster setup, and monitoring.

MCP server that gives LLM agents read-only shell access over SSH. Powers Fawdy (fawdy.com)

A human-like memory for AI Agent

LLM-powered data engineering agent: converts requirement docs and natural language into production-ready SQL, automating data pipeline workflows for ETL and analytics tasks. Open-source, MIT licensed.可嵌入的 LLM 驱动数据开发 Agent：将需求文档和自然语言转换为生产级 SQL，自动化 ETL 和分析任务的数据流水线工作流。开源项目，采用 MIT 协议。

A lightweight Python tracing toolkit for recording execution graphs from existing systems.

Remind Karo - AI assistant and deadline reminder, management app.

HexaFinance is a production-grade financial simulation platform built with Clean/Hexagonal Architecture in Go and a Python (FastAPI + LangGraph) multi-agent AI research service. It offers portfolio management, buy/sell transactions, expense tracking, live market data, and LLM-powered stock research with deterministic fallback.

Electron desktop workspace for orchestrating 5 CLI coding agents (Claude / Codex / Gemini / Kimi / OpenCode) in isolated Git worktrees, with a built-in Sigma Assistant orchestrator that resumes Claude sessions across restarts and observes pane lifecycle events via a mailbox back-channel.

AI Agent Full-Stack Development Roadmap

All your Zotero annotations in one place — filter, export with citations, and turn highlights into linked, searchable ideas. Plus grounded Q&A with clickable page citations.

A multi-agent Claude Code framework for running role-based agents across GTM, product, design, and ops — with plans, project substrate, and a reinforcement loop that compounds learning.

Mod your favourite games without writing code. Modmixer is a desktop app that decompiles/indexes your game and provides your AI coding LLM (Claude, Codex, Kimi, Deepseek etc) with tools to efficiently build and test mods.

自然语言驱动的自动化 SQL 对账系统 — 基于 HelloAgents 多 Agent 框架 · ReAct 推理 · NL2SQL

Universal AI writing skill — tech articles, marketing copy, research reports, Xiaohongshu notes, technical docs. Research-driven, full-pipeline, anti-AI-slop. Bilingual (EN/ZH).

A comprehensive guide and resource collection for building sophisticated agentic AI systems with LangGraph.

The complete AI operating model for software teams — from first idea to production. Three peer-supervised loops (discovery → build → release) over a catalogue of curated packs: skills, subagents, and hooks, each installed in one line. It's npm for your coding agent. Any agent, any stack — Claude Code, Codex, Cursor, Copilot, Gemini, Kiro.

OASiS — an open-source agentic framework for verified multi-physics, multi-code simulation. MCP server letting any AI model drive eight FEM/multiphysics codes (FEniCSx, deal.II, 4C, NGSolve, scikit-fem, Kratos, DUNE-fem, FEBio) plus the SPARTA DSMC particle code, with curated knowledge, verified execution, and cross-code coupling.

Full-pipeline AI agent for Nature-series journal writing. Supports OpenClaw, Claude Code, and Codex. Covers journal selection, drafting, figure generation, citation verification, and reviewer response.

Privacy-first, local-first research notebook for macOS & Windows — chat with your own PDFs, Office docs & web pages and get cited answers, all on your machine. Open-source NotebookLM alternative.

Local-first RAG knowledge base for Pi agents with hybrid code/doc search, reranking, diagnostics, and large-project indexing.

Bring-your-own-key desktop app to build custom AI personas, give them personal knowledge bases, chain them into workflows, and keep conversation memory. Fully local, no subscriptions.

Get insights into token usage for Claude code and Codex sessions with a dashboard and menu bar widget for mac

Web search tool for the pi coding agent — batch queries, ranked readable markdown results ready to fetch. Pairs with pi-smart-fetch.

AI Research Writing Skill (AI论文写作技能) is an agent skill for ML / AI / CV / NLP researchers. Point your coding agent at code, experiment logs, notes, and a venue template; it helps you produce an auditable, evidence-backed LaTeX draft and submission package — not a polished fiction.

AI-Ready Financial Intelligence — 覆盖 CFO 全部工作域的智能体 Skill 体系 | 29 财务场景 × 131 Skills | SAP/Power BI/Bloomberg Connector

One-click migration tool for WorkBuddy — restore conversation history, memory, and MCP connectors after switching accounts. 切换账号后一键恢复对话记录。

Your Hermes agents, one shared brain.

🦅Strix 落地与调教完全指北 —— 这是一个专注于下一代自主 AI 渗透测试代理Strix 的实战部署与性能榨干指南。🎮以及如何通过 --instruction 指令化身“赛博御兽师”操纵 AI 黑客注意力的进阶玩法的核心沉淀。⚡内含国内网络环境下 Docker 0/33 沙箱卡死究极加速方案，💰全网超全的多模型（DeepSeek-V4、Claude、火山引擎、OpenRouter）高性价比接入配置。A comprehensive hands-on guide to maximizing Strix, the autonomous AI agent that thinks and acts like a real hacker.

A reactive Julia notebook built for ambitious work — distributed compute, agentic AI from day one, high-throughput low-latency streaming to the browser, an extension system that integrates anything, and beautiful rendering, publishing, and standalone exports.

Local SQLite memory for LangGraph agents, with provenance, TTL, export, and MCP.

Self-hosted control plane for AI coding agents (Claude Code, Codex, Gemini, Ollama). Shared MCPs, reusable workflows, AI audit that makes any repo AI-friendly. Engineering, not prompting : smaller prompts, fewer hallucinations, lower token bill.

MCP (Model Context Protocol) server for DuckDB — auth and friendly SQL support out of the box

Gatemole compiles human-owned intent into obligations, links evidence artifacts to those obligations, and produces deterministic release decisions for agent-written changes.

List of all the awesome projects into ADK-Hackathon

The project's aim is to develop a chatbot that can help people create spec documents without knowing the specification.To get started with, the bot will consume the spec, JSON schema and serves the user as an expert. So based on a set of questions and answers it will generate an AsyncApi spec document according to the use cases.

A 100% local, privacy-first alternative to Windows Recall — captures your screen, OCRs it, and lets you semantically search and chat with your history using a local LLM. Zero cloud, zero data leakage.

Translate academic papers & monographs (PDF) into a single self-contained bilingual HTML — Chinese/original/side-by-side. Vendor-neutral, works with any file-capable AI agent. 把論文/專著 PDF 翻成單一自包含的中英對照 HTML（純中／對照／純原文三態切換）。廠商中立，任何能操作檔案的 AI agent 都能跑。

Just code, nothing else. A community VS Code chat extension on the Claude Agent SDK

Claude Skill for building clean, professional terminal UIs & CLIs — covers Bubble Tea, Ratatui, Textual, and Ink across Go, Rust, Python, and TypeScript

Cognitive Loop Runtime for Kubernetes

Pier Code is a terminal-native agentic coding tool that understands your codebase and helps you move faster by handling routine tasks, explaining complex code, and managing git workflows through simple natural language commands.

An end-to-end framework for running, sandboxing, and scoring agentic LLMs on complex data-science and econometric replication tasks.

Local-first diff review for AI-assisted development, you have your own judge.

Alpha-stage PRD generation with FastAPI, Streamlit, provider adapters, streamed state, and explicit scope boundaries.

Modular AI-powered testing & monitoring platform — Performance / Test Agent / Test Case Gen / API & UI Automation / Knowledge Base

BossMate — 让 AI 认真读完每个 JD，再决定要不要开口。支持 Codex、Claude Code、OpenCode、Hermes 和 WorkBuddy。

Cross-agent coordination for AI coding agents. One daemon, one database — so multiple agents in one repo can see each other's plan, tasks, and file claims.

Data Wiki turns data into portable knowledge bundles following Open Knowledge Format (OKF) and serves them to AI agents over the Model Context Protocol (MCP).

Durable, searchable memory of your past agent sessions.

Production-ready RAG assistant for Moodle — containerized with Jenkins CI/CD on AWS.

Free AI-powered Amazon product research & marketplace intelligence for 6 platforms. Hidden Profit Index — Sorftime's proprietary algorithm for finding overlooked high-margin products. 80+ MCP tools. Amazon FBA, Walmart, TikTok Shop, Shopee, TEMU, 1688. Claude Code, Codex, Cursor, OpenClaw. Open source.

精选的 Tencent WorkBuddy 学习资料、实战案例、提示词、Skills 与 MCP 资源导航。

Self-hostable sandbox SDK inspired by Cloudflare Sandbox SDK. works on any device

⚡ Multi-Device WhatsApp Bot Engine with Hybrid OpenRouter AI, Modular ESM Plugins, RPG, & E-Commerce Automation

43 Claude Code skills for making content that doesn't read like a machine made it — prose editing with an offline linter, prompt engineering for 40+ image/video/music models with optional one-command execution, and orchestrators from research to a published carousel, reel or proposal. Russian-first, English throughout.

Plasmod: An Agent-Native Database for Cognitive Objects, Event-Driven Materialization, and Structured Evidence Retrieval

多 Agent 协作平台 — supervisor 调度、@ 直连成员、Agentic RAG、Skill 沙箱产出物

PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents

🦾 The Agent Runtime Where Every Thought Is Visible — 7 paradigms, 10 harness shells, zero black boxes.

A 125-task benchmark for knowledge-mediated implicit associations and retrieval blind spots in long-term agent memory.

Skill-guided DevOps agent built on MECE for chat, shell, VPS deployment, Docker, GitLab, AWS EC2

仿微信体验的本地优先 AI 陪伴应用，支持长期记忆、关系成长、朋友圈、群聊与 Android APK。

Give life to an AI agent with a single command.

A structured library of web security flaws for humans and coding agents with vulnerable examples, attack paths, and secure replacements.

AI-powered research and intelligence platform that monitors the web, analyzes information, detects trends, and delivers personalized reports.

VS Code AI 编程助手

ValuSee 是面向消费者的 AI 购物决策与省钱平台，支持商品截图与浏览器扩展采集、SKU 同款及规格识别、跨平台真实到手价核算、评论证据与购买风险分析，并结合预算、已有设备和偏好生成个性化建议；持续提供历史价格、降价监控、保价、退货、保修及买后管理。

mahmud-fca is a Facebook Chat API modified with bug fixes, stability improvements, and extra features for Facebook Messenger bots.

Deterministic, local-only audit reports for Claude Code AI agent sessions. Rust + SQLite. Zero network calls. MCP server for agents.

Automated Unity game development pipeline. Describe a game, get a production ready Unity Project. Powered by Claude Code

🦞 OpenClaw AI Agent Scenario Workshop — visually install, configure & manage OpenClaw agents in 5 minutes. No terminal needed.

locally hosted neural net chat framework with biomimetic RAG (cc0) (by luna+ada) (<3)

AI-powered image editor with LangGraph agent technology. Edit photos using natural language commands through an intelligent conversational interface. Features both traditional manual controls and an AI assistant that understands complex editing requests.

A complete AI-powered video generation pipeline that enhances creative prompts and generates high-quality videos using Google's Veo3 API, requiring only a single API key.

This repository documents my step-by-step learning and implementation of modern MLOps, AI Infrastructure, and Cloud-Native ML systems.

[ACM MM 2026] Slidegen: Collaborative multimodal agents for scientific slide generation

Local, voice-enabled AI assistant on the terminal

轻量化自然语言数据库查询工具，支持多轮对话完善查询目标，精确输出SQL查询语句、执行查询、数据分析并返回自然语言结果。A lightweight system enbales multiple rounds of chat to transfer natural language into SQL query, execute query, perform data analysis and return analysis results.

A self-hosted AI toolkit running locally via Docker Compose, bundling an LLM gateway, workflow automation, and a chat UI — all backed by a shared PostgreSQL database.

Persistent Knowledge Graph Memory for AI Coding Agents. Adds long-term context (goals, strategies, preferences) to Cursor, Windsurf, & VS Code via MCP.

AI agent that watches your errors and opens fix PRs