ワンクリックでManusで任意のスキルを実行

agents-md-guide

スター0

フォーク0

更新日2025年12月26日 05:41

Guide for using and supporting the AGENTS.md standard in VS Code. Use this when asked about AGENTS.md, custom instructions, or repo-level AI agent configuration.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

raphaelmansuy

raphaelmansuy/machine-learning-feature-selection

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

AGENTS.md Guide for VS Code

This skill provides guidance on implementing and using the AGENTS.md standard to provide custom instructions for AI coding agents in VS Code.

Why AGENTS.md?

Standardization: Reduces fragmentation from proprietary files like .cursorrules.
Interoperability: Works across different AI tools (GitHub Copilot, Cursor, etc.).
Efficiency: Saves time by providing structured context (build steps, coding conventions).
Consistency: Ensures AI agents follow project-specific protocols.
Open Standard: Governed by the Agentic AI Foundation (Linux Foundation).

Mental Model

AGENTS.md acts as a centralized instruction manual for AI coding agents at the repo root.

Flow: Repo clone → agent scans for AGENTS.md → parses sections → applies rules during tasks → outputs aligned code.

VS Code Configuration

To enable AGENTS.md support in VS Code:

Enable Setting: Set chat.useAgentsMdFile to true.
Nested Files (Experimental): Set chat.useNestedAgentsMdFiles to true for subfolder instructions.

How to Implement AGENTS.md

Location: Place AGENTS.md at the root of your repository.
Structure: Use clear sections:
- ## Environment: Setup and build instructions.
- ## Coding Style: Linting, formatting, and architectural rules.
- ## Testing: How to run and write tests.
Keep it Concise: Avoid overly verbose rules.

Real-World Scenarios

Open-source Maintenance: AI agents auto-generate PRs following style guides.
Enterprise Code Reviews: Teams use repo-level rules during Copilot-assisted edits.
Indie Dev Prototyping: Automate build and test cycles with tools like Cursor or Codex.

Survival Kit

Day 0: Clone a repo with AGENTS.md; ensure chat.useAgentsMdFile is enabled in VS Code.
Week 1: Create a basic AGENTS.md in a test repo and iterate on sections.
Week 2: Add nested files if needed using experimental settings.

Security & Risks

No Secrets: Never include API keys or credentials.
Goal Hijacking: Be aware that instruction files can steer agent behavior. Review instructions before running autonomous tasks in untrusted repos.

References

このリポジトリの他の Skills

同じリポジトリ

sota-data-cleaning-feature-selection-eda

raphaelmansuy/machine-learning-feature-selection

Master SOTA data prep for Kaggle comps: automated EDA (Sweetviz), cleaning (Pyjanitor), and feature selection (Polars + XGBoost) for medium datasets (100MB–5GB) in Colab.

2025-12-260

skill-creator

raphaelmansuy/machine-learning-feature-selection

Use this skill to create new Agent Skills for GitHub Copilot. It guides you through the process of setting up the directory structure and the SKILL.md file.

2025-12-260

sequentialthinking

raphaelmansuy/machine-learning-feature-selection

Use this skill for systematic problem-solving through structured, iterative thinking. Break down complex problems, iterate on understanding, catch edge cases, and validate solutions comprehensively.

2025-12-260

AGENTS.md Guide for VS Code

This skill provides guidance on implementing and using the AGENTS.md standard to provide custom instructions for AI coding agents in VS Code.

Why AGENTS.md?

Standardization: Reduces fragmentation from proprietary files like .cursorrules.

Interoperability: Works across different AI tools (GitHub Copilot, Cursor, etc.).

Efficiency: Saves time by providing structured context (build steps, coding conventions).

Consistency: Ensures AI agents follow project-specific protocols.

Open Standard: Governed by the Agentic AI Foundation (Linux Foundation).

Mental Model

AGENTS.md acts as a centralized instruction manual for AI coding agents at the repo root.

Flow: Repo clone → agent scans for AGENTS.md → parses sections → applies rules during tasks → outputs aligned code.

VS Code Configuration

To enable AGENTS.md support in VS Code:

Enable Setting: Set chat.useAgentsMdFile to true.

Nested Files (Experimental): Set chat.useNestedAgentsMdFiles to true for subfolder instructions.

How to Implement AGENTS.md

Location: Place AGENTS.md at the root of your repository.

Structure: Use clear sections:

## Environment: Setup and build instructions.
## Coding Style: Linting, formatting, and architectural rules.
## Testing: How to run and write tests.

Keep it Concise: Avoid overly verbose rules.

Real-World Scenarios

Open-source Maintenance: AI agents auto-generate PRs following style guides.

Enterprise Code Reviews: Teams use repo-level rules during Copilot-assisted edits.

Indie Dev Prototyping: Automate build and test cycles with tools like Cursor or Codex.

Survival Kit

Day 0: Clone a repo with AGENTS.md; ensure chat.useAgentsMdFile is enabled in VS Code.

Week 1: Create a basic AGENTS.md in a test repo and iterate on sections.

Week 2: Add nested files if needed using experimental settings.

Security & Risks

No Secrets: Never include API keys or credentials.

Goal Hijacking: Be aware that instruction files can steer agent behavior. Review instructions before running autonomous tasks in untrusted repos.

References

name	agents-md-guide
description	Guide for using and supporting the AGENTS.md standard in VS Code. Use this when asked about AGENTS.md, custom instructions, or repo-level AI agent configuration.