Youtube Transcribe Skill

YouTube Transcript Extraction

Extract subtitles/transcripts from a YouTube video URL and save them as a local file.

Input YouTube URL: $ARGUMENTS

Step 1: Verify URL and Get Video Information

1. Verify URL Format: Confirm the input is a valid YouTube URL (supports youtube.com/watch?v= or youtu.be/ formats).

1. Get Video Information:

• If yt-dlp is available, prefer yt-dlp --get-title "[VIDEO_URL]".
• If using browser automation, extract the title from the page (via snapshot or document.title) for file naming.

Step 2: CLI Quick Extraction (Priority Attempt)

Use command-line tools to quickly extract subtitles.

1. Check Tool Availability:

Execute which yt-dlp.

• If yt-dlp is found, proceed to subtitle download.
• If yt-dlp is NOT found, skip immediately to Step 3.

1. Execute Subtitle Download (Only if yt-dlp is found):

• Tip: Always add --cookies-from-browser to avoid sign-in restrictions. Default to chrome.
• Retry Logic: If yt-dlp fails with a browser error (e.g., "Could not open Chrome"), ask the user to specify their available browser (e.g., firefox, safari, edge) and retry.

```bash # Get the title first (try chrome first) yt-dlp --cookies-from-browser=chrome --get-title "[VIDEO_URL]"

# Download subtitles yt-dlp --cookies-from-browser=chrome --write-auto-sub --write-sub --sub-lang zh-Hans,zh-Hant,en --skip-download --output ".%(ext)s" "[VIDEO_URL]" ```

1. Verify Results:

• Check the command exit code.
• Exit code 0 (Success): Subtitles have been saved locally, task complete.
• Exit code non-0 (Failure):
• If error is related to browser/cookies, ask user for correct browser and retry Step 2.
• If other errors (e.g., video unavailable), proceed to Step 3.

Step 3: Browser Automation (Fallback)

When the CLI method fails or yt-dlp is missing, use browser UI automation to extract subtitles.

1. Check Tool Availability:

• Check if chrome-devtools-mcp tools (specifically mcp__chrome__new_page) are available.
• CRITICAL CHECK: If chrome-devtools-mcp is NOT available AND yt-dlp was NOT found in Step 2:
• STOP execution.
• Notify the User: "Unable to proceed. Please either install yt-dlp (for fast CLI extraction) OR configure chrome-devtools-mcp (for browser automation)."

1. Initialize Browser Session (If tools are available):

Call mcp__chrome__new_page to open the video URL.

3.2 Analyze Page State

Call mcp__chrome__take_snapshot to read the page accessibility tree.

3.3 Expand Video Description

Reason: The "Show transcript" button is usually hidden within the collapsed description area.

1. Search the snapshot for a button labeled "...more", "...更多", or "Show more" (usually located in the description block below the video title).
2. Call mcp__chrome__click to click that button.

3.4 Open Transcript Panel

1. Call mcp__chrome__take_snapshot to get the updated UI snapshot.
2. Search for a button labeled "Show transcript", "显示转录稿", or "内容转文字".
3. Call mcp__chrome__click to click that button.

3.5 Extract Content via DOM

Reason: Directly reading the accessibility tree for long lists is slow and consumes many tokens; DOM injection is more efficient.

Call mcp__chrome__evaluate_script to execute the following JavaScript:

() => {
  // Select all transcript segment containers
  const segments = document.querySelectorAll("ytd-transcript-segment-renderer");
  if (!segments.length) return "BUFFERING"; // Retry if empty

  // Iterate and format as "timestamp text"
  return Array.from(segments)
    .map((seg) => {
      const time = seg.querySelector(".segment-timestamp")?.innerText.trim();
      const text = seg.querySelector(".segment-text")?.innerText.trim();
      return `${time} ${text}`;
    })
    .join("\n");
};

If it returns "BUFFERING", wait a few seconds and retry.

3.6 Save and Cleanup

1. Use the Write tool to save the extracted text as a local file (e.g., .txt).
2. Call mcp__chrome__close_page to release resources.

Output Requirements

• Save the subtitle file to the current working directory.
• Filename format: .txt
• File content format: Each line should be Timestamp Subtitle Text.
• Report upon completion: File path, subtitle language, total number of lines.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: feiskyer
• Source: feiskyer/codex-settings
• License: MIT

Install and usage instructions live in the source repository linked above.