Custom agents

Local MCP hub that lets Claude Code, Codex and Antigravity share one project memory and delegate work to each other on your own subscriptions.

Open-source tool-calling platform for AI agents: 80+ integrations with OAuth, MCP server deployment, memory and agent flows.

Jev for Hermes: skill discovery, multi-skill advice, adaptive reasoning effort, model recommendations, typed assessments, session re-ranking, and computer use.

Use Herdr from anywhere — your iPhone, Android phone or any browser. Watch your coding agents, answer them and run your herdr-projects, privately and securely.

Yandex MCP servers / MCP-серверы для Яндекса: Директ, Метрика, Вебмастер, Wordstat, Трекер, 360, Карты, SpeechKit, YandexGPT, AppMetrica — 10 серверов, 134 инструмента, плюс скиллы Claude Code для SEO-аудита по ЭПОС и проверки спроса.

Free, open-source macOS menu bar app that shows CPU and memory use by app, project, and AI agent

Autonomous security agent on Kali Linux using Qwen 2.5-14b with LM Studio

Pony is the hands, you pick the brain. Let Claude, Grok, Gemini or any MCP agent use your Android phone.

在 dsh 中驱动隐藏 Blockbench：YSM 建模/贴图/动画/无头渲染/游戏验证的闭环工具链（30 tools + Client UI）

会算账的麦当劳点单专家：省钱榜单找最优解 · 一键领券 · 自动用最优券 · 卡路里可视 · 门店导航 · 拼单凑单

Telegram bot development skill for Claude Code and OpenCode: checks the live Bot API docs and the installed library (aiogram, grammY, python-telegram-bot)

Oh My Diff: A Moddable Codex Runtime

Paste this link to your AI agent and say 'setup': it installs the /video-3d command for short voiced 3D character videos.

免越狱将闲置 Kindle（含咪咕版）改造成 AI 实时墨水屏看板：投影 Token Monitor 用量/费用/模型榜 + 全渠道在途 Agent 任务进度条（零依赖标准库 · Agent Skill）

Convert scans, PDFs, TIFFs and whole archives (ZIP/RAR/7z) to Markdown — and see where OCR may have lost text: missing pages, garbage, engine disagreement. Q&A over documents with citations.

AI-powered system design mock interviewer

这篇文章会手把手教你配置OpenClaw+Antigravity，几乎0成本使用openclaw

A terminal dashboard your coding agent sets up: PRs, Jira, calendar and anything else, with Plugins in any language.

Session Recorder by Proofline. Free, open-source Chrome bug reports with annotations, voice, screenshots and one-click GitHub issues.

Evidence over eloquence. Unofficial Claude Code plugin that checks "done" claims and small judgement calls with TypeSafe Jev, and keeps receipts.

Fuli preserves your taste, personality, and judgment across Codex, Claude Code, and other clients supporting MCP and Skills. Your agent picks up where it left off, so you can focus on ideas and decisions—and take a break. As AI learns to do more of the work, Fuli helps you make the transition.

One window for documents, mail, calendar, browser and AI agents. Local-first, MIT.

Self-hosted browser terminal deck for long-running Claude Code and shell agents — every session is a tmux session, so it survives the closed tab, the dropped VPN and a service restart.

Free, open-source MCP connector for Paperclip. Manage agents, issues, approvals, budgets and reports from Claude, ChatGPT, Codex or any MCP app.

Evidence-bounded skill for drafting, rewriting, humanizing, and auditing IEEE/ACM engineering manuscripts.

导演椅 Director's Chair（原名 AutoCrew）：中文创作者的开源 AI 新媒体团队，你只管创作和拍板

MCP server for Android device automation over ADB - screenshots, UI hierarchy, taps, app management, logcat, and screen recording. No companion APK, no telemetry.

Physical desk panel for Herdr, the terminal multiplexer for AI coding agents — shows session state in color and rings a bell when an agent stops to ask for a decision. ESP32-S3 + LVGL, prebuilt firmware.

Zit: a git extension that lets developers and coding agents change one repository at the same time without stepping on each other's toes, and it's kind to your SSD.

The most complete AI-powered JavaScript recon skill - orchestrates katana + waymore + jsluice + hae into one AI agent pipeline. Tools do the breadth, AI does the depth.

AI-agent-powered iOS testing

Human-approved engineering decisions for AI coding agents. Local-first MCP tools for Codex and Claude Code with evidence-based Git diff reviews.

Open-source Apple Health MCP server for the MetricBridge iOS app: query 194 HealthKit metrics from Claude, ChatGPT, Cursor, OpenClaw, Hermes and any AI agent. Zero-dependency, local-first, read-only.

Glyph Engine (formerly prevail-cli): the engine behind Glyph Studio; also a terminal cockpit that asks many AI models at once.

Nerviz — an agent harness for Spring Boot. Claude Code skills, agents and hooks that generate Spring Boot projects and keep AI-written code on-architecture: DDD, dependency direction and naming enforced from domain to tests.

Portable working context for AI coding agents: save your state once, and every agent (Claude Code, Codex, Cursor, Copilot, Gemini CLI) picks it up.

MCP server configs, skill and stdio relay for Zeptap: let AI agents operate a real iPhone from a Mac.

Playwright, but built for AI agents: the browser API models already know, token-efficient a11y snapshots, and scroll-driven collection over the network. Zero-port CDP underneath.

Open-source Computer Use in any agent.

Open-source, Local-first AI (SI) Second Brain.

Minimalist AI-native console for directing agents inside a repository. Tauri + React + Claude CLI.

Question patterns, anti-patterns & calibration notes for typed decision models — Jev & every System One-compatible engine. 60 entries / 10 categories + linter + engine cookbook. EN/中文.

routes agent tasks across herdr panes with composer-safe delegation

فكرتك ← موقع + تطبيق تيليجرام + تطبيق جوال (PWA) مجاناً على Cloudflare، يبنيه لك أي AI. Free website + Telegram Mini App + PWA, built by any AI agent.

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.

Point it at a website, get a grounded chat widget that answers only from that site's pages, with sources. Local crawl and embeddings, Ollama or free Groq/Gemini. Demo a chatbot on a prospect's own content in 10 minutes.

Executable change control for clinical AI: every model update, by an engineer, a pipeline or an AI agent, is checked against its approved change plan (FDA PCCP, EU AI Act Art. 43(4)).

Deterministic AST Context Compressor & 52-Tool MCP Server for Cursor, Claude Code, Antigravity & Windsurf. Reduces prompt tokens by up to 94.6% with 0 hallucinations. TDQS 5.0 Anchor Certified.

Linura — The intelligent system layer for Linux, AI agents, containers, and VMs.

Self-contained HTML slide templates with metadata and AGENTS.md instructions for AI agents to create finished presentation decks.

Open-source control plane for agent workflows

Open-source web scraping CLI. Crawl sites, extract data with CSS selectors or Markdown, and fall back to a proxy or the Decodo Web Data API when blocked. Single Go binary, no runtime or account required.

Pick up in-progress work from one CLI agent inside another: Claude Code, Codex, Grok Build, OpenCode, oh-my-pi, Cursor. Inspired by Grok Build (xAI).

Anti-slop design + GSAP motion skill pack for AI coding agents, with an 82-rule UI detector.

Open problems in math, science and engineering. One number to beat, one program that checks every answer.

Open-source HTML to video / programmatic video framework for AI agents. Free Remotion alternative.

Kubernetes runtime for multi-user DeepSeek Harness: isolated user environments with persistent storage, a Go launcher and a private platform Connector.

Mortiflix Music: make a song with AI, stage by stage (idea, genre, song map, chords, shapes, melody, arrange, feel, export), and take it home as MIDI. Windows and Linux installers; your own Claude Code.

Cross-platform intelligence bridge — correlate Ignition SCADA tags with Studio 5000 PLC logic end-to-end.

Run AI coding agents on your own machines: a local daemon, a relay gateway, and native macOS, iOS and Android clients.

Xiaomi Android TV debloat without root: reversible ADB cleanup, Claude and ChatGPT/Codex guides, FLauncher, DNS ad filtering and low-RAM settings.

Agent Loop Control Plane for coding agents.

Agent-driven video editing with Studio, ComfyUI Apps, CLI and MCP. Turn photos, footage and music into finished edits.

Real Excel files in your web app and your AI agent: an embeddable spreadsheet with a Rust/WebAssembly formula engine, in-place .xlsx saves, and an MCP server.

🌻 Pixel-art AI desktop companion for Windows that sees your screen, helps on her own and remembers your habits · 像素风 AI 桌面伙伴：会看屏幕、主动帮忙、记得你的习惯 · powered by DeepSeek Harness

Free disk space on a macOS dev machine: AI agent worktrees, node_modules, Pods, iOS/Android builds, simulators, emulators, package-manager caches and AI tools data.

An opinionated, repeatable, effortless, and scalable cloud agent toolkit setup for maximizing parallel concurrent long-horizon agentic development and automations.

Your local AI agents talk to each other. You should know what they're saying. A terminal-first guardian that mediates every MCP call, scores it for risk, and asks before anything dangerous happens.

A North Star Metric coach you run inside Claude Code. Audit a metric, decompose what drove it, and build the input tree — grounded in and cited to Amplitude's North Star Playbook.

An SVG canvas that people and AI agents draw on together — diagrams, mockups, posters, whole documents. What you draw is a plain object in your code and plain JSON on disk, so the file itself is where you and the agent meet.

🚀 The first Model Context Protocol (MCP) server for CAXA CAD (数码大方 电子图板) - Control CAXA CAD via AI to automate GB-compliant mechanical drafting and parametric modeling.

Model-invocable context compaction for DeepSeek Harness: mind_map and clear_mind agent tools turn selected history into self-written checkpoints, with auditable GUI tombstones and proactive threshold reminders.

Forge of Thought is an AI cognitive extension: a workshop where your idea is forged into a decision you can stand behind. You and the AI elicit the idea together until it is whole, reviewers that never saw the conversation attack it, and who decided what and why is recorded. Runs in Claude Code.

Curated, vendor-neutral open-source tools, standards, benchmarks and labs for AI Security Engineering, organized by lifecycle (Discover, Design, Build, Break, Defend). From the AI Security Engineers Community.

Open-source AI assistant for your code, files and browser. Native on Windows, macOS and Linux, with Android and mobile web companions. Built with C# and Avalonia.

Free multiple-choice quiz on AI and machine learning, from the basics to research topics. Every option explains why it's right or wrong. English and Spanish.

Catalog of silent failures in RAG systems and AI agents, with reproduce, detect and mitigate code

Every AI coding session on one board, each with its own progress. Claude Code, Codex and OMP in Orca, Terminal, iTerm2, Ghostty or WezTerm (macOS).

Stickman films, directed by your coding agent, rendered as code. No video-generation model, no video API, renders locally.

The data standard that makes your AI coding agent better at data work.

MCP server that lets AI agents drive SOLIDWORKS: parts, sketches, features, assemblies, drawings, export and simulation, with every change verified (C#, .NET 10, official COM API)

Tool-neutral skills and workflows for rigorous AI-assisted development.

Turn rough requests into useful AI assignments. Ambition helps you clarify the outcome, choose evidence and methods, and check the result. Includes examples and a practical prompting guide. Inspired by Felix Rieseberg.

Reviews what your coding agent did, not just the diff: did it do what you asked, what it decided on its own, and what is wrong in the code.

Move conversations between Claude Code, Codex, Cursor, and more. Keep your conversation history, branch into different agents, and compare what happens next.

Local library for Claude Code and Codex skills, prompts, MCP servers, plugins, and environment variables. Desktop app, CLI, and local web UI; delivers each asset into the agent's own workspace or global directory.

Native macOS and Linux app for running many Claude Code and Codex agents side by side — see which one is waiting on you. A GPU-rendered client for herdr.

Agent Skill for bold, research-grounded, non-generic mobile UI (SwiftUI + Jetpack Compose / CMP). Pick a style preset, tune dials, no AI slop.

Dokimi (Greek δοκιμή, "test, proof") measures whether a skill, MCP doc source, agent profile or repository guidance actually helps a coding agent: paired on/off arm comparisons, novel-vs-known content analysis, and repository scans with graded agent runs.

A link-in-bio page that looks hand-made. Pick a look, add your links, publish. Edit it yourself or let your AI do it.

Close the lid, let your agents cook. A tiny tray app that keeps your laptop awake (lid shut, in your bag) only while Claude Code, Codex & friends are working, then lets it sleep. macOS · Windows · Linux.

Prompt-driven workspace and pane management for herdr.

Official implementation of "Local Work, Global Progress: A Convergence Theory of AI Agent Collaboration" — algorithms, theory, and experiments.

Automated Android JNI RegisterNatives recovery, stripped ARM64 .so type reconstruction, OLLVM BR X8 deobfuscator, and universal LLM/MCP bridge for Jadx, Ghidra & IDA Pro 9.0.

Research your family history with an AI agent: parish registers, evidence, a family tree for the Strom app. Runs on your computer.

Orchestrating local agent skills into deterministic workflows. Built by @JiamingShi-JsPhys, Chelate Shell enforces a 4-layer routing architecture for scientific research.

Your coding agents. On your phone. From anywhere.

DIY voice AI agent: an ESP32 that controls your PC, the web, and smart-home devices through a local Python brain. Bilingual (English/Tamil), memory, phone app, mDNS.

An always-on voice AI agent that runs entirely on a Raspberry Pi 5 + AI HAT+ 2. Qwen3 on the Hailo-10H, wake word, Home Assistant, memory, a face. One-script install.

Desktop app for oh-my-pi (omp) on Windows and macOS. Runs the official omp and shares logins, settings and sessions with the terminal version.