tubescribe

Name: tubescribe
Author: AlekseiUL

63stars20forks

How to install tubescribe?

alekseiul/sprut-agent-kit/tubescribe

$npx -y skills add AlekseiUL/sprut-agent-kit --skill tubescribe

Installs into the current project.

Summary

YouTube video summarizer with speaker detection, formatted documents, and audio output.

Files · 1

View on GitHub

SKILL.md

1	metadata:
2	{
3	"claudeclaw":
4	{
5	"emoji": "🎬",
6	"requires": { "bins": ["summarize"] }
7	}
8	}
9	---
10
11	# TubeScribe 🎬
12
13	Turn any YouTube video into a polished document + audio summary.
14
15	Drop a YouTube link → get a beautiful transcript with speaker labels, key quotes, timestamps that link back to the video, and an audio summary you can listen to on the go.
16
17	### 💸 Free & No Paid APIs
18
19	- No subscriptions or API keys — works out of the box
20	- Local processing — transcription, speaker detection, and TTS run on your machine
21	- Network access — fetching from YouTube (captions, metadata, comments) requires internet
22	- No data uploaded — nothing is sent to external services; all processing stays on your machine
23	- Safe sub-agent — spawned sub-agent has strict instructions: no software installation, no network calls beyond YouTube
24
25	### ✨ Features
26
27	- 📄 Transcript with summary and key quotes — Export as DOCX, HTML, or Markdown
28	- 🎯 Smart Speaker Detection — Automatically identifies participants
29	- 🔊 Audio Summaries — Listen to key points (MP3/WAV)
30	- 📝 Clickable Timestamps — Every quote links directly to that moment in the video
31	- 💬 YouTube Comments — Viewer sentiment analysis and best comments
32	- 📋 Queue Support — Send multiple links, they get processed in order
33	- 🚀 Non-Blocking Workflow — Conversation continues while video processes in background
34
35	### 🎬 Works With Any Video
36
37	- Interviews & podcasts (multi-speaker detection)
38	- Lectures & tutorials (single speaker)
39	- Music videos (lyrics extraction)
40	- News & documentaries
41	- Any YouTube content with captions
42
43	## Quick Start
44
45	When user sends a YouTube URL:
46	1. Spawn sub-agent with the full pipeline task immediately
47	2. Reply: "🎬 TubeScribe is processing — I'll let you know when it's ready!"
48	3. Continue conversation (don't wait!)
49	4. Sub-agent notification will announce completion with title and details
50
51	DO NOT BLOCK — spawn and move on instantly.
52
53	## First-Time Setup
54
55	Run setup to check dependencies and configure defaults:
56
57	```bash
58	python skills/tubescribe/scripts/setup.py
59	```
60
61	This checks: `summarize` CLI, `pandoc`, `ffmpeg`, `Kokoro TTS`
62
63	## Full Workflow (Single Sub-Agent)
64
65	Spawn ONE sub-agent that does the entire pipeline:
66
67	```python
68	sessions_spawn(
69	task=f"""
70	## TubeScribe: Process {youtube_url}
71
72	⚠️ CRITICAL: Do NOT install any software.
73	No pip, brew, curl, venv, or binary downloads.
74	If a tool is missing, STOP and report what's needed.
75
76	Run the COMPLETE pipeline — do not stop until all steps are done.
77
78	### Step 1: Extract
79	```bash
80	python3 skills/tubescribe/scripts/tubescribe.py "{youtube_url}"
81	```
82	Note the Source and Output paths printed by the script. Use those exact paths in subsequent steps.
83
84	### Step 2: Read source JSON
85	Read the Source path from Step 1 output and note:
86	- metadata.title (for filename)
87	- metadata.video_id
88	- metadata.channel, upload_date, duration_string
89
90	### Step 3: Create formatted markdown
91	Write to the Output path from Step 1:
92
93	1. `# <title>`
94	---
95	2. Video info block — Channel, Date, Duration, URL (clickable). Empty line between each field.
96	---
97	3. `## Participants` — table with bold headers:
98	```
99	\| Name \| Role \| Description \|
100	\|----------\|----------\|-----------------\|
101	```
102	---
103	4. `## Summary` — 3-5 paragraphs of prose
104	---
105	5. `## Key Quotes` — 5 best with clickable YouTube timestamps. Format each as:
106	```
107	"Quote text here." - [12:34](https://www.youtube.com/watch?v=ID&t=754s)
108
109	"Another quote." - [25:10](https://www.youtube.com/watch?v=ID&t=1510s)
110	```
111	Use regular dash `-`, NOT em dash `—`. Do NOT use blockquotes `>`. Plain paragraphs only.
112	---
113	6. `## Viewer Sentiment` (if comments exist)
114	---
115	7. `## Best Comments` (if comments exist) — Top 5, NO lines between them:
116	```
117	Comment text here.
118
119	- ▲ 123 @AuthorName
120
121	Next comment text here.
122
123	- ▲ 45 @AnotherAuthor
124	```
125	Attribution line: dash + italic. Just blank line between comments, NO `---` separators.
126
127	---
128	8. `## Full Transcript` — merge segments, speaker labels, clickable timestamps
129
130	### Step 4: Create DOCX
131	Clean the title for filename (remove special chars), then:
132	```bash
133	pandoc <output_path> -o ~/Documents/TubeScribe/<safe_title>.docx
134	```
135
136	### Step 5: Generate audio
137	Write the summary text to a temp file, then use TubeScribe's built-in audio generation:
138	```bash
139	# Write summary to temp file (use python3 to write, avoids shell escaping issues)
140	python3 -c "
141	text = '''YOUR SUMMARY TEXT HERE'''
142	with open('<temp_dir>/tubescribe_<video_id>_summary.txt', 'w') as f:
143	f.write(text)
144	"
145
146	# Generate audio (auto-detects engine, voice, format from config)
147	python3 skills/tubescribe/scripts/t

People who liked this also viewed

6 picks

Type

ai-video-generationinference.sh Agent skills for using our API to give your agents access to hundreds of apps and other agentsSkillsJul 2026292k642
imagegen-frontend-webElite frontend image-direction skill for generating premium, conversion-aware website design references. CRITICAL OUTPUT RULE — generate ONE separate horizontal image FOR EVERY section. A landing pageSkillsJul 2026165k66k
hyperframes-mediaWrite HTML. Render video. Built for agents.SkillsJul 2026145k37k
media-useAgent Media OS, the single skill for every media need in a HyperFrames project. Resolve BGM, SFX, image, icon, brand logo, voice, color grade, or LUT into a frozen local file or paste-ready block + leSkillsJul 2026144k37k
general-videoAuthor or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, staticSkillsJul 2026124k37k
product-launch-videoTurn a product or marketing URL, pasted script, or brief into a product launch / promo video — SaaS promos, feature reveals, product demos, app and company launches. Use when the user wants to market,SkillsJul 2026121k37k

tubescribe

63stars20forks

View on GitHub

How to install tubescribe?

alekseiul/sprut-agent-kit/tubescribe

$npx -y skills add AlekseiUL/sprut-agent-kit --skill tubescribe

Installs into the current project.

Summary

YouTube video summarizer with speaker detection, formatted documents, and audio output.

Files · 1

View on GitHub

SKILL.md

1	metadata:
2	{
3	"claudeclaw":
4	{
5	"emoji": "🎬",
6	"requires": { "bins": ["summarize"] }
7	}
8	}
9	---
10
11	# TubeScribe 🎬
12
13	Turn any YouTube video into a polished document + audio summary.
14
15	Drop a YouTube link → get a beautiful transcript with speaker labels, key quotes, timestamps that link back to the video, and an audio summary you can listen to on the go.
16
17	### 💸 Free & No Paid APIs
18
19	- No subscriptions or API keys — works out of the box
20	- Local processing — transcription, speaker detection, and TTS run on your machine
21	- Network access — fetching from YouTube (captions, metadata, comments) requires internet
22	- No data uploaded — nothing is sent to external services; all processing stays on your machine
23	- Safe sub-agent — spawned sub-agent has strict instructions: no software installation, no network calls beyond YouTube
24
25	### ✨ Features
26
27	- 📄 Transcript with summary and key quotes — Export as DOCX, HTML, or Markdown
28	- 🎯 Smart Speaker Detection — Automatically identifies participants
29	- 🔊 Audio Summaries — Listen to key points (MP3/WAV)
30	- 📝 Clickable Timestamps — Every quote links directly to that moment in the video
31	- 💬 YouTube Comments — Viewer sentiment analysis and best comments
32	- 📋 Queue Support — Send multiple links, they get processed in order
33	- 🚀 Non-Blocking Workflow — Conversation continues while video processes in background
34
35	### 🎬 Works With Any Video
36
37	- Interviews & podcasts (multi-speaker detection)
38	- Lectures & tutorials (single speaker)
39	- Music videos (lyrics extraction)
40	- News & documentaries
41	- Any YouTube content with captions
42
43	## Quick Start
44
45	When user sends a YouTube URL:
46	1. Spawn sub-agent with the full pipeline task immediately
47	2. Reply: "🎬 TubeScribe is processing — I'll let you know when it's ready!"
48	3. Continue conversation (don't wait!)
49	4. Sub-agent notification will announce completion with title and details
50
51	DO NOT BLOCK — spawn and move on instantly.
52
53	## First-Time Setup
54
55	Run setup to check dependencies and configure defaults:
56
57	```bash
58	python skills/tubescribe/scripts/setup.py
59	```
60
61	This checks: `summarize` CLI, `pandoc`, `ffmpeg`, `Kokoro TTS`
62
63	## Full Workflow (Single Sub-Agent)
64
65	Spawn ONE sub-agent that does the entire pipeline:
66
67	```python
68	sessions_spawn(
69	task=f"""
70	## TubeScribe: Process {youtube_url}
71
72	⚠️ CRITICAL: Do NOT install any software.
73	No pip, brew, curl, venv, or binary downloads.
74	If a tool is missing, STOP and report what's needed.
75
76	Run the COMPLETE pipeline — do not stop until all steps are done.
77
78	### Step 1: Extract
79	```bash
80	python3 skills/tubescribe/scripts/tubescribe.py "{youtube_url}"
81	```
82	Note the Source and Output paths printed by the script. Use those exact paths in subsequent steps.
83
84	### Step 2: Read source JSON
85	Read the Source path from Step 1 output and note:
86	- metadata.title (for filename)
87	- metadata.video_id
88	- metadata.channel, upload_date, duration_string
89
90	### Step 3: Create formatted markdown
91	Write to the Output path from Step 1:
92
93	1. `# <title>`
94	---
95	2. Video info block — Channel, Date, Duration, URL (clickable). Empty line between each field.
96	---
97	3. `## Participants` — table with bold headers:
98	```
99	\| Name \| Role \| Description \|
100	\|----------\|----------\|-----------------\|
101	```
102	---
103	4. `## Summary` — 3-5 paragraphs of prose
104	---
105	5. `## Key Quotes` — 5 best with clickable YouTube timestamps. Format each as:
106	```
107	"Quote text here." - [12:34](https://www.youtube.com/watch?v=ID&t=754s)
108
109	"Another quote." - [25:10](https://www.youtube.com/watch?v=ID&t=1510s)
110	```
111	Use regular dash `-`, NOT em dash `—`. Do NOT use blockquotes `>`. Plain paragraphs only.
112	---
113	6. `## Viewer Sentiment` (if comments exist)
114	---
115	7. `## Best Comments` (if comments exist) — Top 5, NO lines between them:
116	```
117	Comment text here.
118
119	- ▲ 123 @AuthorName
120
121	Next comment text here.
122
123	- ▲ 45 @AnotherAuthor
124	```
125	Attribution line: dash + italic. Just blank line between comments, NO `---` separators.
126
127	---
128	8. `## Full Transcript` — merge segments, speaker labels, clickable timestamps
129
130	### Step 4: Create DOCX
131	Clean the title for filename (remove special chars), then:
132	```bash
133	pandoc <output_path> -o ~/Documents/TubeScribe/<safe_title>.docx
134	```
135
136	### Step 5: Generate audio
137	Write the summary text to a temp file, then use TubeScribe's built-in audio generation:
138	```bash
139	# Write summary to temp file (use python3 to write, avoids shell escaping issues)
140	python3 -c "
141	text = '''YOUR SUMMARY TEXT HERE'''
142	with open('<temp_dir>/tubescribe_<video_id>_summary.txt', 'w') as f:
143	f.write(text)
144	"
145
146	# Generate audio (auto-detects engine, voice, format from config)
147	python3 skills/tubescribe/scripts/t

People who liked this also viewed

6 picks

Type