菜单
Audit a page set against a named WCAG version and level, classify issues by severity/level, and compute a weighted conformance score. Reach for this on a conformance question.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/mcorbett51090/RavenClaude --skill run-wcag-audit命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | run-wcag-audit |
| description | Audit a page set against a named WCAG version and level, classify issues by severity/level, and compute a weighted conformance score. Reach for this on a conformance question. |
An audit with no named target or that rests on an automated scan alone is not a conformance claim (§3 #1 #2).
WCAG version + level (A/AA/AAA) and the page set in scope — conformance is per-criterion at the target level (§3 #1).
Run the automated scan as a first pass, then manually check the human-judgment criteria it cannot detect (§3 #2).
By severity and WCAG level; one failing Level-A criterion fails the page (§3 #2).
Weighted score + critical-blocker flag via accessibility_calc.py conformance (§3 #1).
A criterion-level audit with severity/level classification, critical blockers, and a weighted conformance score. Traverse Tree 1 in the decision-trees file.