Agent Skills
› mtarcure/claude-vibe-squad
› accessible-media-authoring
accessible-media-authoring
GitHub为图像、视频或音频生成符合无障碍标准的替代文本、带时间轴的字幕及完整转录稿。要求基于源内容创作，禁止虚构，确保功能描述准确且包含非语音提示。
触发场景
需要为生成的媒体素材编写无障碍访问内容
需要将视觉或听觉资产转化为文本形式的辅助信息
安装
npx skills add mtarcure/claude-vibe-squad --skill accessible-media-authoring -g -y
SKILL.md
Frontmatter
{
"name": "accessible-media-authoring",
"audience": "specialist",
"description": "Use when supplied imagery, footage, or sound needs source-faithful access artifacts authored: write functional alt text, time speaker-labeled captions with non-speech cues, and produce a complete transcript; mark decoration explicitly and stop if the asset cannot be perceived. Not for evaluating an interface against accessibility standards."
}
Accessible Media Authoring
Author alt-text, captions, and transcripts for generated or third-party media to accessibility standards.
Steps
- Perceive the asset (multimodal ingest); if it can't be perceived, request it or report
capability_gap— never invent content. - Write alt-text that conveys function/meaning, not a literal pixel description; keep decorative assets marked decorative.
- Author captions with timing, speaker labels, and non-speech sound cues.
- Author a complete transcript covering all spoken and meaningful non-speech audio.
- Verify each artifact against the source asset.
Acceptance
- Missing alt-text/captions/transcript is a PASS-blocker, not a warning.
- Captions are timed and complete; transcript covers all audio.
- No transcript/alt content invented for media that could not be perceived.
版本历史
- d5262e2 当前 2026-09-11 11:15
