LA

leochiu-a/universal-agents

Developer tools

Эта страница переведена автоматически. Технические детали смотрите в оригинале. Открыть оригинал

60 stars Качество 58 Тренд 13MIT

The Lightest Shared Standard for AI Agents

Обзор

A lightweight standard for sharing agent configurations across all AI coding assistants using a single AGENTS.md file. Claude Code sometimes ignores its CLAUDE.md file and doesn't support the AGENTS.md protocol. Use a user memory prompt to force loading: In Claude, pin this via the /memory command so it stays in user memory. Different AI coding assistants use different conventions to organize their configuration files. For example: - Claude Code stores skills in .claude/skills/ /SKILL.md - Cursor stores rules in the .cursor/rules/ folder This fragmentation locks configurations into single platforms. Teams cannot share context and workflow definitions across tools. Universal Agents provides the shared standard: AGENTS.md + .agents/ folder. Any agent—regardless of runtime—reads this one manifest to bootstrap rules and skills. Universal Agents uses a single entry point—AGENTS.md—to describe which skills and rules apply to any task. All skills and rules live in .

README

Показан фрагмент оригинального README. Открыть оригинал.

Universal Agents

A lightweight standard for sharing agent configurations across all AI coding assistants using a single AGENTS.md file.

Quick Start

Run the CLI via npx:

npx universal-agents init           # copy AGENTS.md into your project
npx universal-agents create skill   # scaffold .agents/skills//SKILL.md
npx universal-agents create rule    # scaffold .agents/rules/.md

Apply Universal Agents in Claude Code

Claude Code sometimes ignores its CLAUDE.md file and doesn’t support the AGENTS.md protocol. Use a user memory prompt to force loading:

ALWAYS read AGENTS.md file first

In Claude, pin this via the /memory command so it stays in user memory.

Why Universal Agents?

Different AI coding assistants use different conventions to organize their configuration files. For example:

  • Claude Code stores skills in .claude/skills//SKILL.md
  • Cursor stores rules in the .cursor/rules/ folder

This fragmentation locks configurations into single platforms. Teams cannot share context and workflow definitions across tools.

The Solution

Universal Agents provides the lightest, simplest shared standard: AGENTS.md + .agents/ folder. Any agent—regardless of runtime—reads this one manifest to bootstrap rules and skills.

How it works

Universal Agents uses a single entry point—AGENTS.md—to describe which skills and rules apply to any task. All skills and rules live in .agents/ for version control and easy sharing.

File/Folder Purpose
AGENTS.md Control manifest listing all available skills and rules, with trigger conditions and loading instructions.
.agents/skills/ Reusable task workflows (e.g., code review, testing) — each skill is self-contained and production-ready.
.agents/rules/ Domain guidelines (e.g., API design, React conventions) — long-lived constraints that enforce team standards.

Key advantages

  • One source of truth: AGENTS.md works across Claude Code, Cursor, Codex, and any future AI assistant.
  • Zero boilerplate: No tool-specific file structure. Just read AGENTS.md, follow its protocol.
  • Minimal footprint: Lightweight Markdown—no DSLs, config files, or complex tooling required.
  • Immediate adoption: Agents that understand the AGENTS.md protocol can bootstrap any project instantly.

Execution protocol

  1. Read AGENTS.md first to discover available skills and rules, plus their trigger conditions.
  2. Load skills on demand: when a task matches a skill’s trigger, read the skill file and follow its workflow.
Открыть полный README на GitHub

Рекомендуемые инструменты

Попробуйте другой запрос или уберите фильтр.

Установка

npx skillfish add leochiu-a/universal-agents
Новичок в MCP? Начните с руководства на mcpskill.org →