How do I install the Youtube Transcript skill?

Run npx skillmds@latest add devonjones/youtube-transcript in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Youtube Transcript safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Youtube Transcript?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Youtube Transcript free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Youtube Transcript?

devonjones (@devonjones) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Youtube Transcript by devonjones — Agent Skill

Name: Youtube Transcript
Author: devonjones

Youtube Transcript

Extract transcripts from YouTube videos. Use when the user asks for a transcript, subtitles, or captions and provides a YouTube URL or 11-character video ID. Supports plain text, timestamped output, language selection, and writing directly to a file.

devonjones Updated Sep 22, 2026

File contents

YouTube Transcript

Fetch auto-generated or manually-uploaded captions from a YouTube video via the youtube-transcript-api library (run via uv run --script).

Usage

# Plain transcript to stdout
scripts/get_transcript.py "<URL_OR_VIDEO_ID>"

# With timestamps
scripts/get_transcript.py "<URL_OR_VIDEO_ID>" --timestamps

# Specific language (falls back to default if unavailable)
scripts/get_transcript.py "<URL_OR_VIDEO_ID>" --lang en

# Write directly to a file
scripts/get_transcript.py "<URL_OR_VIDEO_ID>" -o transcript.txt

If the user doesn't specify a filename, either print to stdout or save as <VIDEO_ID>-transcript.txt using -o.

Supported URL Formats

https://www.youtube.com/watch?v=VIDEO_ID
https://youtu.be/VIDEO_ID
https://youtube.com/embed/VIDEO_ID
https://youtube.com/shorts/VIDEO_ID
Raw 11-character video ID

Output Format

Plain (default): one line per caption segment.
Timestamped (--timestamps): [MM:SS] text (or [HH:MM:SS] for long videos).

Handling the Transcript

Never modify the returned transcript content. Do not paraphrase, summarize, or correct. The raw transcript is the source of truth.
For plain output, you MAY reflow lines into paragraphs so sentences aren't cut mid-clause. This is cosmetic only — do not change words.
For timestamped output, preserve the one-segment-per-line structure; do not reflow.

When Captions Aren't Available

The script errors out when a video has no captions at all. Pass that error back to the user — do not fabricate a transcript.

Prerequisites

uv — installs the Python dependencies in an ephemeral venv at invocation time. Install with curl -LsSf https://astral.sh/uv/install.sh | sh or pip install uv.
Network access to youtube.com (the script calls the public timedtext endpoint directly).

---name: youtube-transcriptdescription: Extract transcripts from YouTube videos. Use when the user asks for a transcript, subtitles, or captions and provides a YouTube URL or 11-character video ID. Supports plain text, timestamped output, language selection, and writing directly to a file.---# YouTube TranscriptFetch auto-generated or manually-uploaded captions from a YouTube video via the `youtube-transcript-api` library (run via `uv run --script`).## Usage```bash# Plain transcript to stdoutscripts/get_transcript.py "<URL_OR_VIDEO_ID>"# With timestampsscripts/get_transcript.py "<URL_OR_VIDEO_ID>" --timestamps# Specific language (falls back to default if unavailable)scripts/get_transcript.py "<URL_OR_VIDEO_ID>" --lang en# Write directly to a filescripts/get_transcript.py "<URL_OR_VIDEO_ID>" -o transcript.txt```If the user doesn't specify a filename, either print to stdout or save as `<VIDEO_ID>-transcript.txt` using `-o`.## Supported URL Formats- `https://www.youtube.com/watch?v=VIDEO_ID`- `https://youtu.be/VIDEO_ID`- `https://youtube.com/embed/VIDEO_ID`- `https://youtube.com/shorts/VIDEO_ID`- Raw 11-character video ID## Output Format- **Plain** (default): one line per caption segment.- **Timestamped** (`--timestamps`): `[MM:SS] text` (or `[HH:MM:SS]` for long videos).## Handling the Transcript- **Never modify the returned transcript content.** Do not paraphrase, summarize, or correct. The raw transcript is the source of truth.- **For plain output**, you MAY reflow lines into paragraphs so sentences aren't cut mid-clause. This is cosmetic only — do not change words.- **For timestamped output**, preserve the one-segment-per-line structure; do not reflow.## When Captions Aren't AvailableThe script errors out when a video has no captions at all. Pass that error back to the user — do not fabricate a transcript.## Prerequisites- `uv` — installs the Python dependencies in an ephemeral venv at invocation time. Install with `curl -LsSf https://astral.sh/uv/install.sh | sh` or `pip install uv`.- Network access to `youtube.com` (the script calls the public timedtext endpoint directly).

Frequently asked questions