agentleFS
Community-AccessCommunity-Access/accessibility-agents/.github/skills/media-accessibility/SKILL.md

media-accessibility

Skill409 starsChanged 6 months agoMITView on GitHub

Video/audio accessibility: WebVTT/SRT/TTML captions, audio descriptions, accessible media player ARIA, and WCAG 1.2.x compliance.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No styling support (unlike WebVTT)

    line 69
  2. Distinctive:

    Don't describe: obvious audio cues, subjective interpretations

    line 98
  3. Distinctive:

    Every media player must have keyboard-accessible controls for:

    line 116

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Media Accessibility Skill

Reference data for video, audio, and streaming media accessibility. Used by media-accessibility agent and any web agent encountering <video> or <audio> elements.


WCAG 1.2 Time-Based Media Criteria

SCLevelRequirementTest
1.2.1AAudio-only/video-only: provide text transcript or audio trackTranscript exists alongside media
1.2.2ACaptions for prerecorded audio in videoSynchronized captions present
1.2.3AAudio description OR text alternative for prerecorded videoDescription track or full transcript
1.2.4AACaptions for live audio in videoReal-time captioning active
1.2.5AAAudio description for prerecorded videoDescription track present
1.2.6AAASign language for prerecorded audioSign language interpretation video
1.2.7AAAExtended audio descriptionPaused video for long descriptions
1.2.8AAAMedia alternative for prerecorded videoFull text alternative document
1.2.9AAAAudio-only (live): text alternativeReal-time text transcript

Caption File Formats

WebVTT (Web Video Text Tracks)

WEBVTT

00:00:01.000 --> 00:00:04.000
Welcome to the accessibility course.

00:00:04.500 --> 00:00:08.000
Today we'll cover caption best practices.

00:00:08.500 --> 00:00:12.000
<v Speaker 2>Let's start with file formats.

Key rules:

  • File must start with WEBVTT header
  • Timestamps: HH:MM:SS.mmm --> HH:MM:SS.mmm
  • Speaker identification: <v Speaker Name>Text
  • Maximum 2 lines per caption, 32 characters per line
  • Minimum display time: 1 second
  • Maximum display time: 7 seconds

SRT (SubRip Text)

1
00:00:01,000 --> 00:00:04,000
Welcome to the accessibility course.

2
00:00:04,500 --> 00:00:08,000
Today we'll cover caption best practices.

Key rules:

  • Sequential numbering starting at 1
  • Timestamps use comma for milliseconds (not period)
  • Blank line between entries
  • No styling support (unlike WebVTT)

TTML (Timed Text Markup Language)

  • XML-based, used in broadcast and streaming
  • Supports styling, positioning, and region layout
  • IMSC (Internet Media Subtitles and Captions) is the web profile

Caption Quality Guidelines

MetricRequirement
Accuracy99%+ word accuracy
SynchronizationWithin 1 second of audio
Speaker identificationRequired when 2+ speakers
Sound effectsDescribe in brackets: [applause], [phone rings]
MusicDescribe: [upbeat music], [dramatic orchestral score]
Caption rateMaximum 200 words per minute (3 words/second)
Line lengthMaximum 32 characters per line
Lines per captionMaximum 2 lines
PlacementBottom center default; move for on-screen text

Audio Description

Audio description narrates visual information during pauses in dialogue.

Requirements:

  • Describe: actions, scene changes, on-screen text, facial expressions relevant to plot
  • Don't describe: obvious audio cues, subjective interpretations
  • Timing: fit into natural pauses; for extended descriptions (1.2.7), video pauses automatically
  • Voice: distinct from program audio, clear and neutral

HTML implementation:

<video controls>
  <source src="video.mp4" type="video/mp4">
  <track kind="captions" src="captions-en.vtt" srclang="en" label="English" default>
  <track kind="descriptions" src="descriptions-en.vtt" srclang="en" label="English Audio Descriptions">
</video>

Accessible Media Player ARIA Patterns

Minimum Controls

Every media player must have keyboard-accessible controls for:

  • Play/Pause (role="button", aria-label="Play" / aria-label="Pause")
  • Volume (role="slider", aria-label="Volume", aria-valuemin, aria-valuemax, aria-valuenow)
  • Seek/Progress (role="slider", aria-label="Seek", aria-valuetext="2 minutes 30 seconds")
  • Captions toggle (role="button", aria-pressed, aria-label="Toggle captions")
  • Fullscreen (role="button", aria-label="Enter fullscreen" / aria-label="Exit fullscreen")

Live Region Announcements

<div aria-live="polite" class="sr-only" id="player-status">
  <!-- Announce: "Playing", "Paused", "Video ended", "Captions on", "Captions off" -->
</div>

Keyboard Shortcuts (Media Player Convention)

KeyAction
Space / EnterPlay/Pause
Left ArrowRewind 5 seconds
Right ArrowForward 5 seconds
Up ArrowVolume up
Down ArrowVolume down
MMute/Unmute
CToggle captions
FToggle fullscreen

Transcript Best Practices

  • Provide a full-text transcript adjacent to or linked from the media
  • Include speaker identification, timestamps (optional but helpful), and non-speech audio
  • Transcripts benefit deaf users, search engines, and users who prefer reading
  • Interactive transcripts (click a line to jump to that point) are excellent UX

Live Captioning

  • WCAG 1.2.4 (AA) requires captions for live audio in synchronized media
  • Options: human captioner (CART), AI-powered auto-captioning, hybrid
  • AI auto-captions alone rarely meet the 99% accuracy requirement — human review or correction is recommended
  • WebSocket-based caption delivery for web: push text to a <div aria-live="polite"> container

Common Violations

IssueWCAG SCSeverity
No captions on prerecorded video1.2.2Critical
Auto-generated captions without review1.2.2Serious
No audio description for visual-only content1.2.5Serious
Inaccessible media player controls2.1.1, 4.1.2Critical
No keyboard access to play/pause2.1.1Critical
Autoplay without user control1.4.2Serious
Volume slider not keyboard-operable2.1.1Serious
No transcript for audio-only content1.2.1Critical
Captions poorly synchronized (>2s drift)1.2.2Moderate
Missing speaker identification in captions1.2.2Moderate

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.