Framework
Custom agents
57,747 Custom AI agents indexed on MeshKore — the most complete public catalog, ranked by popularity and updated daily.
57,747 agents · ranked by popularity · refine in the directory →
Custom agents — page 216 of 578
Local-first adaptive tutoring agent foundation reference implementation for iOS and Web, with a civil-service exam application.
🐾 A desktop pet that reacts in real time to your AI agents — Claude Code, Codex, Antigravity, OpenCode, Cursor, Copilot, Gemini.
Semantic memory for AI agents. Local-first, MCP-native.
Local-first skill routing for agent builders — index SKILL.md bundles, get ranked skill plans with confidence and evidence, and serve them to any agent over MCP.
Give any AI agent real hands on your Mac. A stdlib-only Python CLI for volume, brightness, apps, browser, media, screenshots and mic tests. No dependencies, no daemon.
AI proposal-writing & deal-closing engine for Claude Code. Research a prospect, price the engagement (3-tier + ROI), define scope, pre-handle objections & generate client-ready PDF proposals. 14 skills, 5 parallel agents. Auto-builds from your GEO/SEO/reputation audit reports.
SpaceVibe Deck — a minimal macOS terminal for AI agent CLIs
Plan, run, and track AI coding work feature by feature—with local Markdown cards, diagrams and Git worktrees.
An AI code agent that can make patches inside an isolated Docker sandbox
Your team and its agents, in one workspace. Chat, calls, tasks, docs, pull requests and decisions, with Claude Code, Codex, Cursor, Gemini, OpenCode and pi as teammates. Everything links back to the session that did it.
DeepSeek Harness (dsh) 接入飞书/Lark bot,扫码即用:流式卡片、项目工作区、并行任务、多角色 Agent、跨会话通知、对话内模型/密钥管理与安全网守护(dsh 崩溃后飞书仍可自救)。A scan-to-connect bridge bot connecting DeepSeek Harness (dsh) into Feishu/Lark: streaming cards, workspaces, parallel tasks, multi-role agents, cross-session notify, in-chat model/key management, and a safety-net guardian.
Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents
Turn ComfyUI workflows into stable REST APIs and CLI commands. AI agent friendly.
Evidence-driven production readiness checklist for shipping web applications with confidence
Less talk. Finish the work.
💻 Keeps your Mac awake while your AI agents work. Built-in detection for Claude Code, Codex, Cline, Copilot plus presets for Gemini CLI, OpenCode, Aider and Pi.
Text-first browser & background macOS control for DeepSeek Harness (DSH): target the right process and window without taking the user's pointer. 为 DSH 提供文本优先的电脑控制:后台操作 Chromium 与 macOS,不抢前台、不移动鼠标。
Multi-provider Web Search & Fetch for DeepSeek Harness — 8 deeply adapted providers, SearchHints, resilient fallback, and native X / Xiaohongshu retrieval.
Curated Hermes Agent profile packs
用约 600 行纯 Python,从零实现 Cordis / deepseek-harness 的核心 Agent 插件架构:插件系统、Context 服务注册、依赖拓扑排序、事件分发、Waterfall 中间件与可逆副作用。无需真实 LLM API,内置 Fake LLM,通过完整 Agent Loop 直观理解 AI Agent 插件架构。
A lightweight, Git-backed permanent memory for AI agents.
Isometric games made easy — an agentic game kit. 12 agent skills over 9 zero-dependency TypeScript packages. Type one sentence, get a playable game.
Local-first OCR-VLM multimodal document retrieval and independent evaluation prototype.
为AI Agent设计的小红书搜索和总结技能(Skill):提取小红书图文与评论,并由大模型自动合成综合调研报告。 An AI Agent Skill that automates Xiaohongshu (小红书) searching and summarization using Playwright. It extracts multi-modal data (texts, images, comments) and synthesizes a comprehensive markdown report.
Local-first, vault-isolated, AI-native knowledge workspace. Markdown, whiteboards, ER diagrams & mind maps in one editor, with Claude Code / Codex / Gemini CLI agents wired in. Built with Tauri 2 + React.
AI Agent Skill for generating production-quality SVG illustrations, characters & animations. Render-verify-fix loop ensures pixel-perfect output. Works with Claude, Codex, Cursor, Windsurf & any Agent Skills-compatible tool. Rich gradients, five-zone lighting, character construction.
A local-first Python personal AI assistant for reasoning, memory, voice, vision, automation, and device control.
基于多模态 Embedding + Zilliz + Qwen 视觉理解的多模态 RAG 系统。支持 **Cohere / DashScope Embedding** 和 **DashScope / OpenRouter LLM** 双引擎切换。上传 PDF,用自然语言提问,系统自动检索最相关的页面并由 AI 生成回答。 与传统 RAG 不同,本系统**不做文本提取和 OCR**,而是直接将 PDF 页面当作图片处理,通过视觉 Embedding 模型编码,完整保留表格、图表、排版、手写批注等所有视觉信息。
A multi-model CLI coding agent for terminal-first teams
A competitive Cascade game-playing AI agent using adversarial search, ranked 9th out of 206 agents.
Building AI agents and retrieval systems, a masterclass in LLM agents, RAG, and production deployment.
Create Word, Excel, and PowerPoint files (.docx, .xlsx, .pptx) from plain JSON or fully typed APIs — generate, parse, and patch. Built for AI agents, LLM tool-calling, and hand-written code alike; no Microsoft Office required, opens in every major office suite.
统一管理 Claude Code、Codex、Cursor 的 Skill 与 MCP。
Git-native prompt management for AI agent workflows
give your AI agent a spreadsheet it can actually think in: deterministic Excel-compatible math + structured memory (blocks), real .xlsx out. Open source, self-hostable.
Grades the retrieval policies that decide what a long-running agent remembers - comparative scorecards over fixed, inspectable sessions
for show
UI and orchestration for AI coding agents
Combine GPT-6 Astra and Claude Fable 5.1. Astra directs, Claude builds, Astra verifies.
A personal AI agent to help you think, create, and get things done. Powered by open models, on your computer or in the cloud.
doudou-ai-coding 配套的 Agent Skills 仓库。 提供从产品需求分析、原型、数据库设计、开发、测试、部署的全流程 AI 编程支持。
Personal agent skills for design engineering, fluid UI, modern CSS/HTML, and AI judgement
从选题到可复现发布:十个可独立安装的科研 agent skill —— 顶会/期刊论文、科研绘图、基金 proposal 撰写、方法迭代到 SOTA、冷审与代码数据发布。Ten installable agent skills for research: scientific figures, conference and journal manuscripts, research funding proposals, method development, cold review and artifact release.
A secure Windows SSH workspace and MCP server for Codex, Claude Code, Cursor and other AI agents.
Production-grade Kubernetes operator for nousresearch/hermes-agent: declarative spec, security defaults, S3 backups, OCI auto-update with rollback, SSA-based GitOps coexistence, OpenClaw migration.
The Reverse Engineering Toolkit for AI Agents
基于 Java、Spring Boot、Spring AI 和 MCP 构建的可配置化 AI Agent 智能体平台,支持 Flow、Auto、Fixed 多种执行模式、RAG 检索增强与外部工具接入。 A configurable AI Agent platform with Flow, Auto, and Fixed execution modes, MCP tool integration, RAG support, and observability capabilities.
An MCP (Model Context Protocol) server providing LLM agents with direct access to NASA's Common Metadata Repository (CMR).
Agent Skill for creating BioRender-inspired, publication-quality biomedical mechanism figure prompts for GPT Image, Codex, Claude Code, and OpenCode.
An Autonomous Agent for Slay the Spire. Achieved multiple A20H wins with both Ironclad and Defect.
Production-grade AI agent skills for React 19, Next.js 16, TypeScript 7, testing, data, UI, and frontend architecture—ready for Codex, Claude Code, Cursor, Pi, and more.
LibreOffice MCP Server — turn your open documents into tools for Claude, ChatGPT and any MCP client
Ship AI products powered by the best harnesses without building another agent loop
MCP server for Unreal Engine 5.6/5.8 :D token-efficient Blueprint reading, editing, and a persistent project index for AI coding agents
面向外贸公司与工厂的 AI 建站 Skill:按搜索需求规划页面,生成完整静态 HTML,接通询盘邮件。适合产品站、小语种站与 AI 站群(多站运营),支持 Agent 持续更新。
AMT2ABC is an open-source industrial software compiler framework that transforms industrial mechanisms into reusable Atomic Business Capabilities. It models industrial knowledge through AMT and SECP, builds AMT Graphs, and compiles business goals into ABC combinations for apps, agents, and industrial operation loops.
A provider-independent agent SDK with the same query() interface as the Claude Agent SDK, allowing you to use the right model for every turn.
A free and intelligent app for smooth agentic reading and beyond.
A minimal coding-agent harness built to measure what context management actually costs: constraint survival after compaction (8%→83%), default-FAIL verification against false completions, and KV-cache prefix design (21%→90%). Stdlib-only, real deepseek-chat runs
Recognize, transcribe and typeset math formulas locally: natural language / image / broken TeX → valid LaTeX with preview
Navide is an open-source Agentic Development Environment (ADE) that orchestrates AI software teams across the software development lifecycle. Coordinate AI agents, workflows, context, and human collaboration from idea to deployment.
Pi coding agent extension packages — tools, skills, and integrations.
An assistant that lives next to your cursor on Windows. It sees your screen, talks with you, draws on things, solves the math you write, and operates your computer. Its AI runs on your PC, so it's private and works offline.
WorkBuddy Multi-Account Manager & Automated Check-in Toolkit (多账号轮换、全域工作区打通与自动化签到)
Agentic development environment: run Claude Code, Codex, OpenCode and Gemini CLI side by side, and follow every PR from agent to release with its cost. Desktop app for macOS, Windows and Linux. Local, free.
Open-source Acyclic SDK and in-memory reference providers
OPAL AI Agent Skills Collection (Skills.md Compliant)
Coding agent and provider integration patterns covering configuration, SDKs, MCP, REST daemon boundaries, and multi-protocol developer workflows.
Argolink-focused searchable Claude API and agent notes for Claude Opus 5.
A terminal workspace for coding agents: peek into Claude Code, Codex, pi and OpenCode sessions, return to past work, and start work in another project without leaving the dashboard.
Taste-Bench: measuring the long-horizon judgment of LLM agents at real decision forks
Full-duplex Live Voice for Hermes Desktop — GPT-Live-1 on the Codex subscription, with voice tool calls and task delegation to your own agent.
A Multitask "shock absorber" plugin & agent preset for DeepSeek Harness (DSH): the main chat only plans, briefs subagents and summarizes results, while file reads, commands, search and builds all run inside subagent contexts. Four worker types, an approval gate, zero DSH source changes.
Your AI quotas, always in sight. Edge-docked usage rings for Claude, Codex, Gemini, Cursor, Copilot, OpenRouter and more on Windows: live limits, burn rate, history and alerts.
Live2D creation workflow for Codex and AI agents, with PSD/PNG rigging, anime upscaling, Core/WebGL validation, and an example model for VTube Studio.
virtuoso bridge and CLI for AI agent
Agent skills for bespoke websites and app UI: product-derived direction, no generic AI templates, and a Playwright screenshot review loop. Installs into Claude Code, Codex, Cursor, Copilot, Gemini and more via npx.
A Dynamic Island for your Mac that keeps an eye on Codex, Claude Code and DeepSeek Harness: running tasks, output speed and quota pace, right at the notch.
AI Agent workspace for customizing your NBT EVO
Self-improving code for any git repo: give it a measurable goal, a coding agent (Claude Code or Aider) proposes changes, the kernel scores them, keeps winners and rolls back failures. Open-source AlphaEvolve / FunSearch-style evolution loop, sandboxed, with an auditable ledger.
Build, deploy, and connect AI agents to SAP Joule via the A2A protocol on BTP Cloud Foundry — directly from Claude Code. Supports TypeScript (Express or CAP) and Python agents with LangGraph and SAP GenAI Hub.
Build products with AI agents. Map the architecture, edit real screens, and send changes to your existing coding chat. Local and open source.
Turn intent into work. Open-source Desktop/Web agent harness with long-horizon memory, live evaluation and recoverable batches.
Rebuild any UI from a reference and prove it with a score. An Agent Skill for Claude Code and other coding agents: captures a site, screenshot or screen recording, then scores the clone on layout, style, pixels and motion until it passes.
Beautiful screen recordings, made by your favorite AI agent.
An intelligent shopping website that combines a multi-agent framework with personalized recommendation, interactive assistance, and product exploration.
Give Your AI Agents Access to Quantum Compute
Your $0/month personal AI agent, running 24/7 on GCP's free tier. One-click deploy. Zero cost. Forever.
This is a simple AI assistance for Docusaurus generated documentation. It is a simple chatbot that can answer questions about your docs.
Ship software with AI that knows your codebase, your standards, and gets better every sprint.
An unofficial .NET API wrapper for CharacterAI (https://character.ai/)
A Flutter implementation of Google's Dialog Flow improved
A Collection of Pydantic Models to Abstract IRL
Stage 1: Sensitive Email/Chat Classification for Adversary Agent Emulation (espionage). This project is meant to extend Red Reaper v1 which was presented at RSA San Francisco 2024.
Code for NeurIPS2023 accepted paper: Counterfactual Conservative Q Learning for Offline Multi-agent Reinforcement Learning.
Framework to create chatbots on all platforms and on the browser - https://newbot.io
Comprehensive evaluation of 15 AI coding agents (Cursor, Copilot, Claude, Replit, v0, Warp, etc.) with implementations, screenshots, and professional scoring. Published on Turing Post.
Empower Your Productivity with Local AI Assistants
Create a GPT chatbot for any GitHub repo in just 30 seconds