# Youtube Processor

> Process YouTube videos into summarized Obsidian notes. Use when given a YouTube URL to summarize, extract insights, or turn videos into notes. Triggers on "summarize this video", "process this YouTube", "what's this video about", or any YouTube URL shared for processing.

- **Type:** Skill
- **Install:** `agentstack add skill-nicepkg-ai-workflow-youtube-processor`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [nicepkg](https://agentstack.voostack.com/s/nicepkg)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [nicepkg](https://github.com/nicepkg)
- **Source:** https://github.com/nicepkg/ai-workflow/tree/main/workflows/content-creator-workflow/.claude/skills/youtube-processor
- **Website:** https://ai-workflow.xiaominglab.com/

## Install

```sh
agentstack add skill-nicepkg-ai-workflow-youtube-processor
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# YouTube Processor

## What This Does

Takes a YouTube URL, extracts the transcript, and you (Claude) summarize it. Outputs Obsidian-ready markdown. Zero friction: share a link, get actionable notes.

## When to Use

- "Summarize this video: [URL]"
- "Turn this YouTube into notes"
- "What's this video about?"
- "Process this for my newsletter"
- Any YouTube URL shared for processing

## Location

This skill's tools live at:
```
/Users/eddale/Documents/GitHub/powerhouse-lab/skills/youtube-processor/tools/
```

## How It Works

**Step 1:** Python extracts the transcript (no API key needed)
**Step 2:** You (Claude) summarize using your intelligence + context
**Step 3:** You write the Obsidian-formatted output

This approach means you can use mission-context, newsletter-coach, and other skills during summarization.

---

## Instructions

### Which Method to Use

| Environment | Method |
|-------------|--------|
| **Claude Code** | Local Python script (Step 1a) |
| **Claude.ai / Mac Client** | API via WebFetch (Step 1b) |

---

### Step 1a: Extract Transcript (Claude Code)

Run the Python tool to get the transcript:

```bash
cd /Users/eddale/Documents/GitHub/powerhouse-lab/skills/youtube-processor/tools && \
python3 get_transcript.py --url "[URL]"
```

For JSON output (easier to parse):
```bash
python3 get_transcript.py --url "[URL]" --json
```

### Step 1b: Extract Transcript (Claude.ai / Mac Client)

Use the API endpoint via WebFetch:

```
WebFetch: https://youtube-processor-eight.vercel.app/transcript?url=[VIDEO_URL]
```

The API returns JSON:
```json
{
  "success": true,
  "video_id": "abc123",
  "language": "en",
  "transcript": "...",
  "char_count": 5000,
  "word_count": 850
}
```

**Example prompt for WebFetch:**
"Extract the transcript text from the response"

---

### Step 2: Summarize the Transcript

Once you have the transcript, summarize it based on what the user needs:

**Quick Summary:**
- Headline (1 sentence)
- Key points (3-5 bullets)
- Main takeaway

**Detailed Analysis:**
- Headline summary
- Key points with context
- Main takeaways
- Action items mentioned
- How this relates to Ed's work (if relevant)

**Newsletter Mining:**
- Hook ideas for an article
- Core insight/framework
- Story beats for anecdotes
- Takeaways for readers
- Newsletter angle for The Little Blue Report

### Step 3: Format for Obsidian

Create markdown with this structure:

```markdown
---
source: YouTube
video_id: [ID]
url: [URL]
processed: [YYYY-MM-DD HH:MM]
tags: [youtube, video-notes]
---

# [Video Title or Topic]

**Link**: [URL]
**Processed**: [Date]

## Summary

[Your summary here]

---

## Full Transcript

[The transcript]

---

_Generated by youtube-processor skill_
```

### Step 4: Save (Optional)

Save to Ed's Zettelkasten:
```
/Users/eddale/Documents/COPYobsidian/MAGI/Zettelkasten/
```

Filename format: `YT - [Topic] - YYYY-MM-DD.md`

---

## Error Handling

| Error | Meaning | What to Do |
|-------|---------|------------|
| "Transcripts disabled" | Creator turned off captions | Video can't be processed |
| "No transcript found" | No English captions available | Try a different video |
| "Video unavailable" | Private, deleted, or age-restricted | Check the URL |

---

## Examples

### Example 1: Quick Summary

**User says:**
```
Summarize this: https://www.youtube.com/watch?v=VIDEO_ID
```

**You do:**
1. Run: `python3 get_transcript.py --url "https://www.youtube.com/watch?v=VIDEO_ID"`
2. Read the transcript output
3. Provide a summary to the user

### Example 2: Save to Obsidian

**User says:**
```
Turn this video into notes and save it: [URL]
```

**You do:**
1. Extract transcript with the Python tool
2. Summarize the content
3. Format as Obsidian markdown
4. Use Write tool to save to Zettelkasten
5. Confirm: "Saved to [filepath]"

### Example 3: Newsletter Mining

**User says:**
```
I want to write about this video for the newsletter: [URL]
```

**You do:**
1. Extract transcript
2. Analyze for newsletter angles (hooks, insights, story beats)
3. Present the angles
4. Offer to hand off to newsletter-coach skill

---

## Integration Points

- **newsletter-coach**: After extracting video insights, hand off for article development
- **mission-context**: Use Ed's voice and style when summarizing
- **task-clarity-scanner**: Action items from videos can be added to daily notes

---

## Dependencies

The Python tool requires:
```
youtube-transcript-api
```

Install if needed:
```bash
pip3 install youtube-transcript-api
```

---

## Version History

| Version | Date | Changes |
|---------|------|---------|
| 1.1 | 2026-01-03 | Added Vercel API for Claude.ai/Mac client support |
| 1.0 | 2026-01-02 | Initial build with transcript extraction |

## Notes & Learnings

- youtube-transcript-api works without API keys
- Most videos have auto-generated English captions
- Claude doing the summarization is better than Python calling the API (can use context)
- ~3-5 second transcript extraction for typical videos

## Source & license

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

- **Author:** [nicepkg](https://github.com/nicepkg)
- **Source:** [nicepkg/ai-workflow](https://github.com/nicepkg/ai-workflow)
- **License:** MIT
- **Homepage:** https://ai-workflow.xiaominglab.com/

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

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-nicepkg-ai-workflow-youtube-processor
- Seller: https://agentstack.voostack.com/s/nicepkg
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.