Custom agents

Lab content for enterprise workshops: Databricks Genie agents, AWS data lakes, full-stack AWS, and LLMOps. Datasets, exercises, and trainer keys.

Hermes Second Brain 🧠. Obsidian second-brain vault template for a Hermes AI agent — session archive + memory review pipeline, Kanban boards, templates, Canvas pipeline map, Dataview queries, and cross-platform (Windows/Linux/macOS) install scripts.

AiCode 的 miuix/HyperOS 风格 UI 二改版（基于 jieapi/AiCode，GPL-3.0）

agent-harness-engineering-pattern

Local-first AI workspace with a real Linux desktop inside it.

A native iPhone, iPad and Mac app for the Hermes Agent you run (beta). hermesapp.dev

CLI that lets coding agents search and read other agents' session logs and transcripts (Claude Code, Codex, Copilot CLI). Read-only, no network, pure Go.

🔍 Design details for coding agents to build, compare and follow changes: layout, SVG assets, bundle checks, screenshot comparisons and saved-revision tracking. Reads Figma .fig files offline. CLI, MCP and host plugins.

Text tasks to your Grok Bot. It reaches your Mac over Tailscale, where Claude Code does the job, adding other agents (Codex, Cursor and more) as needed. Nothing is merged, pushed, sent or spent until you say yes.

JSON AI Studio — AI assistant for JSON: edit, fix and explain JSON in plain English, with every change as a reviewable diff.

The 100% Safe-Rust Embedded Quad-Model AI Database & Cognitive Memory Engine (SQL, Vectors, GraphRAG, Documents)

Evidence-grounded incident investigation for code repositories, built with hybrid retrieval, PostgreSQL/pgvector, and Gemini.

Independent verifier for trading strategies written by AI agents and humans: catches look-ahead bias, hidden costs and overfitting. MCP server, CLI, GitHub Action, signed certificates.

Deploy Agent Skills from git repos, declaratively, from a single pollen.yaml

Ask a technical question and get well researched answer with citations.

AI writes. SPARDA proves. A deterministic, offline gate that catches when an AI edit removes a guard, exposes a route, or breaks an invariant — no API key, right in the agent edit loop.

Automatically repair broken Playwright E2E tests. When a UI change renames or restructures an element and a test's selector breaks, the engine diagnoses the failure, patches the broken selector/wait, verifies the new selector against the live DOM, then re-runs the test until it passes (or a retry cap is hit) and writes the fix back

Call any MCP server's tools from bash, so AI agents load a tool schema only when they need it.

Directory-scoped Laravel CLAUDE.md rules — mirrors laravel/laravel structure

Teach your AI coding agent what your DevExpress XAF application actually does. Roslyn-based extraction of entities, controllers, business rules and Model Editor customizations.

⚒️ Forge is a native macOS terminal multiplexer built for parallel CLI agents and long-running tasks, powered by libghostty.

A self-evolving AI Agent Skill (agentskills.io standard) that turns Claude Code, Cursor, Windsurf, or OpenCode into a personalized X (Twitter) social media manager. Features automated onboarding, voice-pattern extraction, algorithm-optimized posting/replying, and self-improving memory logs.

Play three.ws Coin Clash — the community faction war backed by real holdings + pump.fun data — from any AI agent. Read the live battle board and leaderboard (public), or enlist your wallet and rally power for a faction (Solana-signed writes).

Manage file-based identity credentials with CredAgent (Alibaba Cloud's open-source credential protection software), allowing access only by authorized programs/scripts and preventing AI agents or unauthorized programs from leaking credentials.

Research-first scouting agent skill for dependency, API, SaaS, prior-art, and build-versus-buy decisions before writing code.

💡A vault of project policy for AI agents. You talk; the agent writes, links docs, and recalls. CLI + MCP.

Delegate tasks from Claude Code to Google Antigravity CLI (agy) in headless mode. Save your Claude quota across macOS, Win & WSL.

Smooth sailing for your multi-agent CMUX world

Health check for Claude Code skill trees: usage, dead skills, and conflicts

ビジネス要求から保守運用までの成果物を一貫生成するエンタープライズ向けAI駆動開発フレームワーク。Claude Code用Skills・テンプレート・成果物フロー・ガードレールなど、大規模開発に必要な要素を網羅的に収録。

一会话，一扇门，一界命数。诸天万界DSH：DeepSeek Harness 上的文字修罗场。十九扇门点封面启程，不助手不提纲只写正文；誊出来的，是你活过的天书。

Hearting — the agent setting for Claude Code, Codex, and OpenCode. One portable contract projected onto all three, with routed capabilities, sealed cross-harness dispatch, per-node model tiers, and a live Fleet view. Formerly agent_setting.

Meridian — the open-source Arc browser alternative for macOS, Windows and Linux, built for AI agents (MCP server, Claude, coding-agent workbench)

AI running coach for Garmin: syncs Garmin Connect to your computer, explains HRV, readiness and VO2max, puts workouts on your watch. Free MCP server for Claude.

Open data toolbox for QGIS. Explore live satellite imagery, weather, elevation and OpenStreetMap data with a few clicks, and let Claude, Cursor, VS Code and other AI agents drive QGIS through a permission gated MCP server with an approval queue and audit log.

The open-source generative UI engine. A model writes a JSON spec, Ripple renders it as a live interface while it streams.

Unified Marketing Analyst agent skill designed to integrate Google Ads, Analytics, and Merchant Center data.

ucws-hackathon-agent

Access Z.ai coding models within GitHub Copilot Chat

Статический анализ выгрузок конфигураций 1С:Предприятие: индексирует метаданные, управляемые формы и BSL-модули в машиночитаемый контекст для LLM-инструментов. Python-библиотека и CLI.

VisualRAG is a multi-modal AI pipeline that lets you: Index images — YOLOv8 detects objects; CLIP encodes each image into a 512-d vector stored in FAISS Query with text — Ask any natural language question; CLIP retrieves the most visually similar images; Zephyr-7B generates a grounded answer

Claude and Qwen editions of Dynamic Workflows from Scratch

把 Muse（Meta 个人 AI 助理）接进微信：腾讯官方 iLink 协议、长轮询秒级收消息、连发消息合并、主动推送、媒体断点续传

interactive, model checkable specs

Free, open-source AI assistant for macOS. Talk about your screen, get spoken step-by-step guidance and visual answers, and delegate tasks to specialist mates. Native SwiftUI; bring your own Claude, ChatGPT or compatible API.

A self-hosted software factory of Claude Code agents with review, QA and a judge merge gate.

Give every page a soul.

Community SKILL scenarios for mirroir-mcp — AI-driven automation flows for real iOS devices

One local-first desktop workspace for every AI coding agent — Codex, Claude, OpenCode and your own models.

A DSH plugin debugging toolkit. Your agent inspects, diagnoses, and fixes its own runtime. | DSH 插件排障工具箱。让 Agent 自检自修身处的运行时。

DSH for Android, built on LittleWhale (dsh ported to Android by Miuzarte)

通过微信管理本机多个 Pi 会话的 Pi 扩展. 独立后台 daemon, 前后台会话自动交接, 支持远程下发任务与上下文回顾.

AI-native CAD engineering agent that turns natural language and technical documents into structured 2D drawings and validated DXF files.

Video editing, by talking. A local browser video editor that Claude edits with you.

网络工程师助手：华为/新华三/锐捷/迪普/思科 五厂商命令知识库，40 篇结构化文档，可检索、跨厂商对照、高危提醒

Local MCP bridge connecting ChatGPT to OpenClaw, with browser and desktop automation, file operations, project search, command/process control, dynamic tool discovery, and a local call monitor.

随机灵感词 × 外部 agent 模拟研究者的 TUI 循环

A pixel-art open space where your scheduled jobs and AI agents come to work. A shadcn/ui component for Next.js and React.

Project task CLI for agents (stdlib + git)

A live 3D office for the bots in your Grok Bot desktop app: pets that doze, work and wave when they need you. Idea from @0xCodila. Not affiliated with xAI.

One entry point for Claude Code over pinned open-source Blender skill libraries: registry, workflows, profiles, single Blender MCP server

Auto-pilot for agents

Private local memory for any AI agent: Markdown database, SQLite search, MCP, zero runtime dependencies.

MCP server for AI agents: high-fidelity web search (Exa) + tiered web fetch (Exa→browser→Firecrawl) with SSRF guard. Drop-in replacement for built-in WebSearch/WebFetch in Claude Code, Codex, Grok, Cursor and all other agents.

Ask Key: the Mac key manager for AI agents. API keys stay encrypted on your Mac; Claude Code, Codex and Cursor find them over MCP, you approve each use, and the value goes only to the program, never to the AI. | 请旨：给 AI 用的 Mac 密钥管家。AI 自己找得到密钥，普通密钥直接放行，敏感密钥由你审批，密钥只交给目标程序，AI 看不到明文。

Engine-first browser for AI agents. One Rust binary, built-in V8, no Chromium — HTTP is the only protocol face.

Working files for the Vibe Testing course — logs to investigate, an epic to refine, an ng-apimock server, and a Postman collection. For use with QAcart-Premium/ship-app.

Open-source instrumentation for AI coding agents. Capture activity, concurrency and working patterns across Claude Code, Codex, Cursor and more.

Claude OSINT Skill

Product Wiki is an open-source harness for building software with coding agents, especially useful for non-technical builders who want to build complex products without the codebase turning messy.

Browse the x402 bazaar, decode 402 payment challenges, and pay per call for paid APIs and MCP tools in USDC or $THREE on Solana and Base, without leaving VS Code. Powered by three.ws.

The source code for the Oblivion package and experimentation pipeline corresponding to the "Oblivion: Self-Adaptive Agentic Memory Control through Decay-Driven Activation" paper.

木铎 MUDUO · 人类传播与社群增长引擎 | Evidence-aware communication methodology Skill for AI agents: 216 mechanism nodes · 13 platform adapters · long-form intelligence · 每条建议带证据等级与适用边界 · by 谁是专家

Agent skills to police AI-generated frontend work for design system conformance.

A runtime learning loop for coding agents. Turns Claude Code, Codex, Cursor, Antigravity and model-router runs into human-approved, measured changes to the agent harness: guard rules, skills, sub-agents and routes. It tunes the harness, not the model.

Point at the model. Let the Agent read what you meant.

Telegram + Discord alert bots for Robinhood Chain memecoins — new launches, graduations, whale trades. Hosted service, free + premium tier.

MCP server and companion skill for AI agents to build, edit, inspect and export native parametric SOLIDWORKS models.

AI 写好了软件，授权怎么保护？从公开破解案例反推卡密、激活码、离线许可证与联网验证设计。Vendor-neutral Agent Skill for software licensing, license key validation, offline activation, device binding & anti-tampering reviews.

Make your AI show its work. Yantra turns a description, from you or your coding agent, into interactive step-by-step diagrams, answer pages and books: laid out automatically, validated, and easy to share. CLI, MCP server, playground and Python SDK.

Decentralized AI-to-AI communication — Let Claude Code, Codex, and OpenCode talk to each other. Local, LAN, or across the internet with E2E encryption.

x402 payment rail for USDG on Robinhood Chain — spec-conformant exact/EIP-3009 middleware, client fetch wrapper, self-hostable facilitator.

Interactive AI concepts lab — hands-on demos from LLM fundamentals to agent engineering, built on a director-style demo framework. 互動式 AI 教學實驗室，每個概念都能親手玩。

Local MCP server and Chrome browser control runtime that lets AI agents use your existing browser session.

Curated list of the best AI apps and tools, organized into 10 categories. 127 verified, hand-picked picks, updated regularly.

Rift DS: an AI-ready React design system. Accessible components on composable tokens, with light/dark theming and CSS-variable overrides.

share a live agent session with a link

A multi-agent system that searches for news, analyzes sentiment, creates social media posts, and generates promotional posters using AI.

A free bilingual AI engineering roadmap with notes, code examples, and learning resources.

A complete Claude Code plugin for YouTube pre-production: 16 subagents and an orchestrator with approval gates — research, ideas, scripts written in your own learned voice, thumbnails, SEO, metadata, Shorts, APA-cited research reports and monetization. Any niche, any language.

키움증권 REST API CLI — 주식 시세·계좌·주문·자동매매·미국주식을 터미널에서 (Kiwoom Securities CLI for Korean & US stocks)

pstack (poteto's engineering methodology) ported to oh-my-pi. Model-agnostic omp marketplace: /marketplace add negoro26/pstack-omp

Agentic-on-the-Edge smart home implementation w/Microsoft Agent Framework & Model Context Protocol

给 AI 助手的求职搜索技能：中国校招、实习、社招与招聘日历。自然语言提问，获取岗位要求、来源链接与投递入口；支持 Skill / MCP。

Installer for Ziva.sh, an AI assistant inside the Godot editor that edits scenes and scripts, reads editor errors and output, runs tests, fixes failures, and playtests games.

VisuallyJs releases and issue tracking

TypeScript GraphRAG on PostgreSQL/pgvector. Answers carry the evidence they came from, or return an explicit abstention. MCP server, hosted read tier, x402 paid judgments.

把仓库丢给你的 AI Agent，一句话把 Unity 游戏上架 Google Play / 微信小游戏 / TapTap：从注册开发者账号到提交审核，步骤来自真实上架。Agent skills that ship your Unity game.

Local MCP bridge that lets an AI assistant help build tabletop battle maps from assets you already own.

Framework-agnostic frontend design skill for UI/UX — CSS layers, OKLCH tokens, container queries, WCAG 2.2, Design.md