Custom agents

AI Challenge for 30 days to learn and implement AI in Snowflake Intelligence platform

A native macOS app that runs Claude Code as a city of robot-staffed offices. Every project is a building, every floor a team, and every animation a real CLI event.

Agent Skill collection: photo-press-v1 turns a photo into a printed-poster via 12 real printmaking crafts (woodcut, ukiyo-e, watercolor, letterpress, risograph, silkscreen, cyanotype...). 照片印坊：照片→工艺海报，风格是工艺不是滤镜。

An egress gate for CLI coding agents. Masks secrets and customer data in the request body before it reaches the model provider, and the agent keeps working.

Live map of your local GitHub Copilot agents in the Copilot app, Copilot CLI, and VS Code: sessions, sub-agents, tools, data flows, and models. macOS menu bar app, Windows tray app, and web UI. Not an official GitHub product.

Open-source, self-hosted control plane for sandboxes: one API, tenancy, quotas and policy-gated placement across OpenSandbox, 15 more provider adapters (E2B, Daytona, Vercel, AWS, Azure, Google Cloud…) or your own backend.

Build a complete, secure SaaS end to end with your CLI agent: 22 lifecycle skills, prevention-first security, RLS audit, real payments — and field notes from production apps that passed every test. Pairs with Superpowers.

AI-powered penetration-testing pipeline. Specialised agents recon, exploit and adversarially verify web-application vulnerabilities, then produce evidence-backed reports and per-finding fix branches. White-box (source) and grey-box (no-source) modes. Built by CERT-EU.

Free desktop app to run Claude Code, Codex, Cursor, OpenCode and other AI coding agents in parallel, each in its own git worktree. macOS, Windows, Linux. Steer every agent from your phone. A Conductor alternative.

Agent skills that teach AI agents how to work with MikroTik RouterOS

Zero-trust multi-tenant semantic memory for Google ADK, enforced in PostgreSQL with pgvector and Row-Level Security.

ADIF MCP server — AI-agent access to Amateur Radio logging per ADIF 3.1.7

A newspaper your printer prints every morning: a Claude routine writes it overnight from your notes, GitHub Actions lays it out and checks every page, a Raspberry Pi Pico W prints it.

Tiny realtime kanban board for AI agents and the humans watching them

Open-source test impact analysis for agent-driven CI

A collection of tested CLI coding agents for Termux on Android, featuring installation guides, setup instructions, and usage documentation.

AI second brain vault for Obsidian — an agent compiles web clippings into a cited Markdown wiki. Claude Code + Git, model-neutral. 옵시디언 기반 AI 지식관리(PKM) 템플릿

⚔️ AI 代码审查 Skill —— 自动挑刺、定级、给下一轮指令，让 agent 交活前自己把关。

Your agent writes the signature and the failing tests. You get to finish the function. A Claude Code plugin.

Autonomous AI agents inside a Qubes-isolated sandbox - tag-scoped Admin API access with dom0-mediated trust boundary.

GUI client for omp, the oh-my-pi coding agent, on this computer and remote machines over SSH and Tailscale

A native macOS terminal for the AI era: tabs that show which agents are working, done or waiting for you, plus an IDE-style project sidebar.

21 skill agent untuk engineering dan UI. Satu halaman per skill, gerbang verifikasi, nol klaim kosong.

A collection of my conference talks, slides, demos, and resources from events.

Code and dataset for 'ToolAlignBench: Investigating Alignment Conflicts in Tool-Calling Enabled LLMs' — do safety-aligned tool-calling agents obey deployment instructions? (ICML 2026 Pluralistic Alignment Workshop)

Personal LLM gateway for saving expensive tokens

Watch your Claude Code (and Codex) agents live as pixel characters: handoffs, cost, loops, stuck permissions, context pressure — and familiars that evolve with good habits.

Fraud workflow: millisecond GuardianAI screening plus LangGraph multi-agent investigation.

Fully local, offline autonomous multi-agent software enterprise framework powered by Ollama.

A local-first, provider-agnostic platform to benchmark, score, and diagnose autonomous AI coding agents on real-world polyglot software tasks.

Offline agentic RAG security assistant on Raspberry Pi 5 + Hailo-8L NPU. Real-time person detection, re-identification, and natural-language querying fully local, no cloud, no API costs.

A self-hosted workspace where a small team and its AI agents build software together, on one Linux server you own. Claude Code sessions, shared projects, tasks, services and native apps.

Two-tier patient monitoring: Kafka event streaming, deterministic safety rules, and a LangGraph + RAG reasoning agent with Qdrant retrieval and benchmarks.

Build better tests. Run what matters. Remember what worked. An open-source test intelligence and quality engineering layer.

Your AI pet that lives in the menu bar.

dsh（DeepSeek Harness）的 QQ 群聊/私聊通道插件，基于 OneBot v11（NapCat）。 把 QQ 的世界映射进 Agent 上下文，把行动权交给 Agent 本身——插件不做任何代答。

为「SnowLuma + qq-bridge + DSH（大肥鱼）+ DeepSeek」这条 QQ 机器人链路补上插件层：50 个插件 / 84 个工具，覆盖群管、签到、记账、备忘、长期记忆、语音合成、AI 画画、成语接龙等小游戏。网页点一下安装就变成 AI 可调用的 MCP 工具，零依赖、一个命令启动、完全开源——装完机器人真能替你办事。One-click MCP plugin market for AI QQ bots: DeepSeek · SnowLuma · qq-bridge · OneBot · 50 plugins / 84 tools.

把 朗读桥（dsh-tts-bridge）念出来的声音，变成真正的微信语音条，发到当前微信聊天

Production-grade Corrective RAG (CRAG) & Hybrid Retrieval engine combining BM25 lexical search, Dense Vectors, Reciprocal Rank Fusion (RRF), and ONNX Cross-Encoder reranking.

A lightweight dynamic memory engine for LLM Agents. Core: Decay weights + Spatiotemporal Graph + Explainable Retrieval. No Docker, No Vector DB, Just SQLite.

Scout MCP server: web search, scraping, and extraction tools for Claude Code, Codex, Gemini CLI, Antigravity, Cursor, and any MCP client

Production-oriented RAG pipeline — PDF upload, FAISS retrieval, Groq LLM (free), custom evaluation metrics, Langfuse observability. Deployable on Hugging Face Spaces.

OntoAgents — a personal work & learning assistant grounded on ontologies: build AI agents and their knowledge bases (ontology lifecycle: build → run → audit), fully self-hosted & local. Go · CloudWeGo Eino/ADK · React · SQLite · MCP

Production-ready sustainability compliance app featuring multimodal packaging extraction, deterministic rules engine, and real-time Gemini AI assistance.

A free, structured guide for the engineering manager who needs to learn AI fast. Eleven modules take you from LLMs and RAG to hiring, evaluation and strategy, with 100 AI interview questions and the templates you need to build and manage AI teams. No ML background required.

Experimental RAG study assistant for answering questions and generating learning materials from PDFs.

Enterprise AI RAG Assistant built with Hugging Face, Sentence Transformers, FAISS, and Streamlit. Supports semantic search, vector retrieval, and context-aware question answering using Retrieval-Augmented Generation (RAG).

RAG platform built from scratch - async pipelines, 5 chunking strategies, RAGAS evaluation, Celery workers, PostgreSQL lifecycle management

An AI agent framework for Python.

Have Claude, ChatGPT (Codex), and Gemini answer independently and discuss autonomously in the same group. Once they reach a consensus, have them give you a team conclusion. Run it locally using your own subscriptions.让 Claude、ChatGPT(Codex)、Gemini 在同一个群里独立作答、自主讨论，达成共识后给你一份团队结论。本地运行，用你自己的订阅

A privacy-first, open-source Retrieval-Augmented Generation platform with real-time confidence metrics and source highlighting

A local knowledge base of your own documents served to Claude over MCP

herdr skill: fan one task out to N parallel coding agents — Claude Code, OpenCode, Codex, Gemini & 20+. Same-kind spawn, workspace-locked, keep-fold-for-review. Agents never commit.

A SimCity-style city run by 5 GMI Cloud AI mayors and ~100 AI residents. Godot 4, built with Factory AI Droid.

Principal AI Engineer & Forward-Deployed AI Architect | Enterprise & Solution Architecture | Agentic AI, Multi-Agent Systems, RAG & MCP | MLOps/LLMOps, Evaluation, Observability & Responsible AI | Customer-focused discovery, hands-on delivery, measurable business outcomes. PRIME: build, measure, operate.

Is Agent Code Less Maintainable Than Human Code?

An agent that reads an unstructured document (an invoice, expense claim, or contract), extracts structured data, decides what action to take, and proposes it — but cannot execute anything without explicit human approval.

An LLM-animated daemon bound to Systemd. Forges intelligent agents using Litestar, Pydantic AI & Graph, displayed on a Svelte frontend. Supports many Compositions.

Self-hosted remote control of your own computers from Claude and ChatGPT. Files, terminal and processes on any machine you pair — through a relay you host, on your domain.

Assistant vocal français pour Windows : interface holographique 4K, mémoire, second cerveau (Claude Code), Gmail. Édition améliorée non officielle de Mark LV (FatihMakes).

Assistente desktop para Windows com Electron, Python, memória local e automação assistida por UFO².

A framework for AI characters that live with you, on your own machine.

Pomelu: open-source Android & iOS app for Hermes Agent (Nous Research). Chat with your self-hosted AI agent on phone, foldable and tablet: voice, files, approvals, Bot Screen.

Self-hosted Cloudflare Tunnel management platform

Acceptance testing for the agentic era.

A personal agent that lives on your machine, remembers what matters, gets better as it works, and hands the heavy work to the best agent for it.

GitHub 官方 MCP Server 的 Operit 接入配置（远端直连，46 个工具）

An LM Studio plugin that turns a local model into a coding agent: files and shell, memory, git and GitHub, web search and a real browser. No API keys, nothing leaves your machine.

A local Model Context Protocol (MCP) server integration to search actor filmographies, match movie soundtrack albums, and generate curated Spotify playlists.

AI research agent by MattVidPro: ask a question and it searches the web, reads the pages and writes up the answer. Quick mode and deep-research mode in a simple web app. Built on Bright Data search and scraping plus any LLM.

Mira Voss — a terminal AI agent on npm, and her world: art, docs, and her stories in English and Chinese.

Evidence-first resume crafting, JD tailoring, professional packaging, review, and verified PDF delivery for Codex.

A pi (@earendil-works/pi-coding-agent) provider for @evenrealities/even-terminal — run the pi coding agent on Even Realities G2 smart glasses + R1 ring.

Self-hosted OpenMuse: a personal AI agent whose conversations live in your own Postgres (or embedded PGlite), not a hosted chat cloud. Alpha, MIT.

WhoBot AI (呼波特) — Enterprise AI Phone Digital Employee Platform | 企业级 AI 电话数字员工平台

AI slide generator for coding agents: a Claude Code and Codex skill plus CLI that turns any content into one self-contained, editable HTML slide deck (single file, zero dependencies): 115 templates, charts, diagrams, logos, brand type roles, in-file editor, vector PDF, verified layout. Open-source alternative to PowerPoint, Gamma and Slidev.

Send coding tasks to an AI coding agent on your own machine from Telegram. Watch it work in real time, approve risky actions, and get every change as a clean, reviewable git branch. Multi-agent (Claude Code, Gemini, Codex, Aider), cross-platform, written in Go.

Open-source Agent Skills for DOT / FMCSA compliance. Drop in any AI coding agent (Claude Code, Cursor, Cowork) so it knows trucking regulations like a fleet safety pro.

Give Claude video generation through Muapi MCP — setup, model guidance, and API examples

Modular Agentic Genesis Intelligences

Remote Linux workspaces over SSH for DeepSeek Harness — the agent's bash/read/write/edit/glob/grep tools transparently run on your server. Zero remote install.

BondLens: evidence-first Chinese bond analysis agent — deterministic tools, optional LLM narration under guardrails, Trust Layer, live/snapshot/static data, Docker/CI.

DeepSeek Harness Desktop 一键安装器：富江人格预设（工作助手）+ 240 个随包技能。Windows / .NET Framework 4.8 / AnyCPU，只改 DSH_HOME，不动程序目录。

DeepSeek Harness（DSH）插件：对话内右键 Agent 文件，预览图片和视频缩略图，打开、定位、复制路径与引用。

启文: An AI storywriting agent on Android mobile platform, designed by 401A-Coding of School of Software, Tsinghua University

Curated links for learning, using, extending, and deploying Hermes Agent

AI map editor for GTA V: describe a scene and a Claude agent builds it in the real game world, checks it and exports a FiveM, alt:V, RAGE:MP or Menyoo map (.ymap).

A clean Android AI agent built on pi and Termux. 一个基于 pi 和 Termux 的简洁安卓 AI Agent。

500+ research, web search, scraping & data extraction APIs for AI agents. Pay per call with USDC via x402.

See what your coding agents actually do and spend: sessions, context and cost for Claude Code, Codex, OpenCode, DeepSeek Harness, WorkBuddy. Local-first.

3D objects management and visualization

Super-agent for Claude Code: an open-source Router–Worker multi-agent system after DeepSeek's Super-agent design. One personal agent that runs many Claude Code sessions for you.

A code-first AI agent for your terminal. Durable sessions, Mission Control, voice, and a headless daemon you steer from your phone or desktop. Bring your own model.

CONCISE - Cuts prose filler

DSH 插件：接管 DeepSeek Harness 的会话压缩，产出结构化交接摘要、可回查分段与检索工具｜A DSH plugin that takes over session compaction.

Jupiter Perps (SOL/BTC/ETH) bot: your SI agent calls the shots within your rules

Open-source AI application engineering platform for building business-safe AI applications from existing systems or new business models—with governance, auditability, and private deployment built in. Existing System / New Business → Protocol → AI Application → Business-safe Agent → Governance → Audit → Private Deploy

Autonomous CTF solver — HackTheBox + FlagYard, agent-based

QQ 私聊里的 AI 家教：拍照解题、作业批改、错题复习、个性化出题、定时上课

한국 세법 근거 MCP — 국세청 해석·판례·통칙·집행기준·시점별 조문·조세조약, 주장 대조·보기 판정 | Korean tax law MCP server: NTS rulings, court & tribunal cases, statutes as in force on a date, treaties, statement-vs-statute checks