refero-skill-methodology
暂无简介
- License 许可
- MIT
- 最近更新
- 2026-10-09
- Stars
- ★ 2
- Forks
- 1
03 SKILL.md 正文
Refero Skill Methodology
Source: http~/.seosona/path/ Date Ingested: 2026-06-12
1. Core Philosophy
The refero_skill introduces a strictly enforced "Research-first methodology" for AI Agents. It completely bans generic AI UI generation.
- No design from vibe memory: Every major visual, layout, content, or interaction decision must trace to a concrete reference or a craft rule.
- Anti-Averaging Quality Gates: Do not average multiple references into a "safe, generic middle". When references conflict, choose one dominant direction and preserve its sharp, distinctive traits.
- Anti-AI-Slop: Avoid current generic AI design defaults (e.g., unjustified "calm editorial" looks, generic olive/clay/terracotta palettes, or decorative serif word swaps) unless specifically requested or justified by product context.
2. Bundled Craft Knowledge
When the local reference pack is installed at 2_KNOWLEDGE/frameworks/refero_skill/references/, the repository contains deep, senior-level craft knowledge. The agent should read these before specific implementations:
typography.md: Senior typography rules.color.md: Semantic color token usage.motion.md: Micro-interactions and animations.anti-ai-slop.md: A strict checklist to prevent cheap-looking AI outputs.copywriting.md: Conversion-focused UX writing.craft-details.md: Forms, focus states, and accessibility.
3. SEOSONA OS Directive
Whenever SEOSONA OS generates a UI component, website page, or design system for Website SEOSONA, it MUST act as a Senior Product Designer.
It must first define tokens (type scale, colors, spacing) and review the anti-ai-slop.md guide to ensure the final output is distinctive, premium, and fully escapes the generic AI "vibe".
04 同仓库更多 Skills
web_crawler_katana
High-speed web crawling using Katana patterns for deep site mapping, URL discovery, and content extraction.
server_log_analyzer
Analyzes Nginx/Apache server logs to identify Googlebot crawl patterns, errors, and performance bottlenecks.
schema_markup_generator
Auto-generates JSON-LD structured data for any page type (Article, Product, FAQ, LocalBusiness, etc.).
internal_linking_optimizer
Analyzes and optimizes the internal link graph of a website for SEO authority distribution.
google_trends_analyzer
Detects trending topics and seasonal patterns using Google Trends data for content planning and keyword strategy.
brand_context_analyzer
Tracks brand mentions, sentiment, and online reputation across web sources.