An AI agent is a software system that autonomously perceives its environment, makes decisions, and takes actions to achieve goals — typically powered by large language models (LLMs). Unlike simple chatbots, AI agents can use tools, access external data, execute code, browse the web, and manage multi-step workflows without continuous human guidance.
AI agents range from coding assistants that edit files and run tests (Claude Code, Gemini CLI, Cursor) to browser automation systems (browser-use), computer-use agents (CUA), workflow builders (Activepieces), and multi-agent teams (CrewAI, deer-flow). The ecosystem includes memory layers (mem0), frontend toolkits (CopilotKit), sandboxed execution environments (E2B, Daytona), and full productivity platforms (Cherry Studio, LobeHub).
MCP server for XNS S3-compatible storage — agent-driven Relayer install & management. Mirrored from GitLab.
Research-backed prose registers for AI coding agents. Six levels from classy humanized (liner) to bare keywords (morse). Concise output that stays clear.
Short hair, short code. A Claude Code plugin for the simplest code that works: up to 82 % less code on Opus, every safety check kept.
Unofficial, community-built Peach Payments integration skill for AI coding agents (SKILL.md): discovery-first, plugin-first, with verified webhook, result-code, and amount-unit patterns.
Benchmark lending rates for Bitcoin DeFi, read from contract state on Stacks, Base and Ethereum. Published daily, free to use, built for agents.
WebJs is an AI-first full-stack JavaScript web components framework. You get production-ready architecture from your very first prompt.
Switch sessions. Keep building. Git-based AI delivery workflow for humans and AI sessions: context in project files, handoffs across models, tools, sessions and people, and a build-verify-review-fix loop that stops at your merge decision.
Declare what done means — hivekit builds until it's true. Terraform for software: HIVE.md intent-as-source, reconciled by a deterministic Planner → Generator → Evaluator loop.
24/7 Midnight build lab — Compact starters, Lace kit, Agent Escrow (Cardano partner chain)
Local-first Markdown notes where you delegate edits to an AI agent and review them hunk by hunk.
在 dsh 里装上这个插件即可，无需登录、注册或填 API Key，就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费，不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, DeepSeek V4.1 Flash and Kimi K3 among them. Completely free, with no usage cap.
Yui hub: roadmap, progress, Yui Lines spec + playground. Lives at yuigui.com
Shared memory for AI agents, no database, no server, just a GitHub repo. Memories that worked get stronger; memories that failed fade.
You already have the agent. This is everything around it. great_cto runs Claude Code as a pipeline of 70 specialist agents — an independent model checks each stage before the next builds on it, spending caps refuse rather than warn, and three decisions stay yours: what gets built, how, and whether it ships.
Trust policy layer for AI agents, MCP servers, skills, and plugins.
The security-first skill manager for AI agents — every install runs a security scan. Manage skills & MCP servers across 90 agents. Zero-dependency CLI.
Open-source, self-hosted alternative to Airtable, Fibery, Notion databases and ClickUp: a database and work management platform built for AI agents, agent-first (Claude Code, Codex, Cursor, any MCP client). Tables, relations, formulas, rollups, workflows, markdown/PDF documents, one SQLite file. REST, CLI, MCP server. Zero dependencies, MIT.
Local-only MCP server that drives the real Chrome you already have open (your profile, your logins, your session) through a companion extension. 45 tools across navigation, perception, capture, and CDP-backed trusted input that pages receive with isTrusted=true.
The open-source, end-to-end presentation platform - build, present, publish, and collaborate. Self-hosted, MCP-native, fully themeable.
The Project Operating Layer for AI Coding Agents