project-guides

Author and review standards-compliant guide.md files for AI coding assistants.

926|71|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/GoogleChrome/modern-web-guidance-src --skill project-guides
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-guides
Source: https://github.com/GoogleChrome/modern-web-guidance-src/tree/main/.agents/skills/project-guides
Command: npx skills add https://github.com/GoogleChrome/modern-web-guidance-src --skill project-guides

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the problem of inconsistent, incomplete, or poorly structured web platform guidance that causes AI coding assistants to generate legacy, incorrect, or non-performant implementation code. It standardizes the end-to-end authoring and review process for guide.md files, which are the sole source of truth for real-world coding agents retrieving modern web guidance.

Core Features & Use Cases

  • Standardized Guide Authoring: Enforces strict formatting, content, and self-containment rules for guide.md files to ensure they are fully actionable for AI agents without external references.
  • Mandatory Compatibility & Fallback Requirements: Requires all guides to include up-to-date browser Baseline status, fallback strategies for older browsers, and restrictions on unsafe polyfills to ensure broad, safe compatibility.
  • Eval Infrastructure Guidelines: Provides clear rules for creating supporting eval infrastructure (demo.html, expectations.md, grader.ts) to validate that guides produce correct agent implementations.
  • Use Case: A team building a modern web guidance library for AI coding assistants uses this Skill to ensure all guides follow consistent standards, so agents reliably generate modern, performant, and accessible implementation code for web platform features.

Quick Start

Use the project-guides skill to author or review a new guide.md file for a modern web platform feature, following all mandatory formatting, content, and fallback requirements.

Frequently Asked Questions about project-guides

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create web platform guidance for AI coding assistants that prevents legacy code generation?

To create web platform guidance for AI coding assistants, author self-contained guide.md files that serve as the sole source of truth. This enforces strict formatting and content rules to ensure agents generate modern, performant implementation code.

What is the best way to standardize feature documentation for AI coding agents?

Standardizing feature documentation for AI coding agents requires enforcing consistent formatting, mandatory fallback strategies, and browser Baseline status alignment within self-contained guide.md files. This prevents agents from generating incorrect implementation code.

How do I include browser compatibility fallback strategies in AI web guidance?

Include browser compatibility fallback strategies in AI web guidance by mandating up-to-date browser Baseline status and restricting unsafe polyfills within your guide.md files. This ensures broad, safe compatibility across older browsers.

Do I need eval infrastructure to validate AI coding agent guides?

Yes, you need eval infrastructure to validate AI coding agent guides. Create supporting files like demo.html, expectations.md, and grader.ts to confirm that your guide.md files produce correct agent implementations for web platform features.

Why do my AI coding assistants generate incorrect web platform implementation code?

AI coding assistants generate incorrect web platform implementation code when guidance is inconsistent, incomplete, or poorly structured. Standardizing the authoring process for self-contained guide.md files eliminates external dependencies and prevents legacy code generation.

Can I use external references inside guide.md files for AI coding agents?

No, you cannot use external references inside guide.md files for AI coding agents. Guides must be fully self-contained and actionable without external dependencies, ensuring AI agents reliably implement modern web platform features.