learn-this

Name: learn-this
Author: michalparkola

498stars65forks

How to install learn-this?

michalparkola/tapestry-skills/learn-this

$npx -y skills add michalparkola/tapestry-skills --skill learn-this

Installs into the current project.

Summary

Unified content extraction and action planning. Use when user says "learn-this <URL>", "learn this <URL>", "weave <URL>", "help me plan <URL>", "extract and plan <URL>", "make this actionable <URL>", or similar phrases indicating they want to extract content and create an action

Files · 1

View on GitHub

SKILL.md

1	# Tapestry: Unified Content Extraction + Action Planning
2
3	This is the master skill that orchestrates the entire Tapestry workflow:
4	1. Detect content type from URL
5	2. Extract content using appropriate skill
6	3. Automatically create a Ship-Learn-Next action plan
7
8	## When to Use This Skill
9
10	Activate when the user:
11	- Says "learn-this [URL]" or "learn this [URL]"
12	- Says "weave [URL]"
13	- Says "help me plan [URL]"
14	- Says "extract and plan [URL]"
15	- Says "make this actionable [URL]"
16	- Says "turn [URL] into a plan"
17	- Provides a URL and asks to "learn and implement from this"
18	- Wants the full Tapestry workflow (extract → plan)
19
20	Keywords to watch for: learn-this, learn this, weave, plan, actionable, extract and plan, make a plan, turn into action
21
22	## How It Works
23
24	### Complete Workflow:
25	1. Detect URL type (YouTube, article, PDF)
26	2. Extract content using appropriate skill:
27	- YouTube → youtube-transcript skill
28	- Article → article-extractor skill
29	- PDF → download and extract text
30	3. Create action plan using ship-learn-next skill
31	4. Save both content file and plan file
32	5. Present summary to user
33
34	## URL Detection Logic
35
36	### YouTube Videos
37
38	Patterns to detect:
39	- `youtube.com/watch?v=`
40	- `youtu.be/`
41	- `youtube.com/shorts/`
42	- `m.youtube.com/watch?v=`
43
44	Action: Use youtube-transcript skill
45
46	### Web Articles/Blog Posts
47
48	Patterns to detect:
49	- `http://` or `https://`
50	- NOT YouTube, NOT PDF
51	- Common domains: medium.com, substack.com, dev.to, etc.
52	- Any HTML page
53
54	Action: Use article-extractor skill
55
56	### PDF Documents
57
58	Patterns to detect:
59	- URL ends with `.pdf`
60	- URL returns `Content-Type: application/pdf`
61
62	Action: Download and extract text
63
64	### Other Content
65
66	Fallback:
67	- Try article-extractor (works for most HTML)
68	- If fails, inform user of unsupported type
69
70	## Step-by-Step Workflow
71
72	### Step 1: Detect Content Type
73
74	```bash
75	URL="$1"
76
77	# Check for YouTube
78	if [[ "$URL" =~ youtube\.com/watch \|\| "$URL" =~ youtu\.be/ \|\| "$URL" =~ youtube\.com/shorts ]]; then
79	CONTENT_TYPE="youtube"
80
81	# Check for PDF
82	elif [[ "$URL" =~ \.pdf$ ]]; then
83	CONTENT_TYPE="pdf"
84
85	# Check if URL returns PDF
86	elif curl -sI "$URL" \| grep -i "Content-Type: application/pdf" > /dev/null; then
87	CONTENT_TYPE="pdf"
88
89	# Default to article
90	else
91	CONTENT_TYPE="article"
92	fi
93
94	echo "📍 Detected: $CONTENT_TYPE"
95	```
96
97	### Step 2: Extract Content (by Type)
98
99	#### YouTube Video
100
101	```bash
102	# Use youtube-transcript skill workflow
103	echo "📺 Extracting YouTube transcript..."
104
105	# 1. Check for yt-dlp
106	if ! command -v yt-dlp &> /dev/null; then
107	echo "Installing yt-dlp..."
108	brew install yt-dlp
109	fi
110
111	# 2. Get video title
112	VIDEO_TITLE=$(yt-dlp --print "%(title)s" "$URL" \| tr '/' '_' \| tr ':' '-' \| tr '?' '' \| tr '"' '')
113
114	# 3. Download transcript
115	yt-dlp --write-auto-sub --skip-download --sub-langs en --output "temp_transcript" "$URL"
116
117	# 4. Convert to clean text (deduplicate)
118	python3 -c "
119	import sys, re
120	seen = set()
121	vtt_file = 'temp_transcript.en.vtt'
122	try:
123	with open(vtt_file, 'r') as f:
124	for line in f:
125	line = line.strip()
126	if line and not line.startswith('WEBVTT') and not line.startswith('Kind:') and not line.startswith('Language:') and '-->' not in line:
127	clean = re.sub('<[^>]*>', '', line)
128	clean = clean.replace('&', '&').replace('>', '>').replace('<', '<')
129	if clean and clean not in seen:
130	print(clean)
131	seen.add(clean)
132	except FileNotFoundError:
133	print('Error: Could not find transcript file', file=sys.stderr)
134	sys.exit(1)
135	" > "${VIDEO_TITLE}.txt"
136
137	# 5. Cleanup
138	rm -f temp_transcript.en.vtt
139
140	CONTENT_FILE="${VIDEO_TITLE}.txt"
141	echo "✓ Saved transcript: $CONTENT_FILE"
142	```
143
144	#### Article/Blog Post
145
146	```bash
147	# Use article-extractor skill workflow
148	echo "📄 Extracting article content..."
149
150	# 1. Check for extraction tools
151	if command -v reader &> /dev/null; then
152	TOOL="reader"
153	elif command -v trafilatura &> /dev/null; then
154	TOOL="trafilatura"
155	else
156	TOOL="fallback"
157	fi
158
159	echo "Using: $TOOL"
160
161	# 2. Extract based on tool
162	case $TOOL in
163	reader)
164	reader "$URL" > temp_article.txt
165	ARTICLE_TITLE=$(head -n 1 temp_article.txt \| sed 's/^# //')
166	;;
167
168	trafilatura)
169	METADATA=$(trafilatura --URL "$URL" --json)
170	ARTICLE_TITLE=$(echo "$METADATA" \| python3 -c "import json, sys; print(json.load(sys.stdin).get('title', 'Article'))")
171	trafilatura --URL "$URL" --output-format txt --no-comments > temp_article.txt
172	;;
173
174	fallback)

People who liked this also viewed

6 picks

Type

learn-this

498stars65forks

View on GitHub

How to install learn-this?

michalparkola/tapestry-skills/learn-this

$npx -y skills add michalparkola/tapestry-skills --skill learn-this

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Tapestry: Unified Content Extraction + Action Planning
2
3	This is the master skill that orchestrates the entire Tapestry workflow:
4	1. Detect content type from URL
5	2. Extract content using appropriate skill
6	3. Automatically create a Ship-Learn-Next action plan
7
8	## When to Use This Skill
9
10	Activate when the user:
11	- Says "learn-this [URL]" or "learn this [URL]"
12	- Says "weave [URL]"
13	- Says "help me plan [URL]"
14	- Says "extract and plan [URL]"
15	- Says "make this actionable [URL]"
16	- Says "turn [URL] into a plan"
17	- Provides a URL and asks to "learn and implement from this"
18	- Wants the full Tapestry workflow (extract → plan)
19
20	Keywords to watch for: learn-this, learn this, weave, plan, actionable, extract and plan, make a plan, turn into action
21
22	## How It Works
23
24	### Complete Workflow:
25	1. Detect URL type (YouTube, article, PDF)
26	2. Extract content using appropriate skill:
27	- YouTube → youtube-transcript skill
28	- Article → article-extractor skill
29	- PDF → download and extract text
30	3. Create action plan using ship-learn-next skill
31	4. Save both content file and plan file
32	5. Present summary to user
33
34	## URL Detection Logic
35
36	### YouTube Videos
37
38	Patterns to detect:
39	- `youtube.com/watch?v=`
40	- `youtu.be/`
41	- `youtube.com/shorts/`
42	- `m.youtube.com/watch?v=`
43
44	Action: Use youtube-transcript skill
45
46	### Web Articles/Blog Posts
47
48	Patterns to detect:
49	- `http://` or `https://`
50	- NOT YouTube, NOT PDF
51	- Common domains: medium.com, substack.com, dev.to, etc.
52	- Any HTML page
53
54	Action: Use article-extractor skill
55
56	### PDF Documents
57
58	Patterns to detect:
59	- URL ends with `.pdf`
60	- URL returns `Content-Type: application/pdf`
61
62	Action: Download and extract text
63
64	### Other Content
65
66	Fallback:
67	- Try article-extractor (works for most HTML)
68	- If fails, inform user of unsupported type
69
70	## Step-by-Step Workflow
71
72	### Step 1: Detect Content Type
73
74	```bash
75	URL="$1"
76
77	# Check for YouTube
78	if [[ "$URL" =~ youtube\.com/watch \|\| "$URL" =~ youtu\.be/ \|\| "$URL" =~ youtube\.com/shorts ]]; then
79	CONTENT_TYPE="youtube"
80
81	# Check for PDF
82	elif [[ "$URL" =~ \.pdf$ ]]; then
83	CONTENT_TYPE="pdf"
84
85	# Check if URL returns PDF
86	elif curl -sI "$URL" \| grep -i "Content-Type: application/pdf" > /dev/null; then
87	CONTENT_TYPE="pdf"
88
89	# Default to article
90	else
91	CONTENT_TYPE="article"
92	fi
93
94	echo "📍 Detected: $CONTENT_TYPE"
95	```
96
97	### Step 2: Extract Content (by Type)
98
99	#### YouTube Video
100
101	```bash
102	# Use youtube-transcript skill workflow
103	echo "📺 Extracting YouTube transcript..."
104
105	# 1. Check for yt-dlp
106	if ! command -v yt-dlp &> /dev/null; then
107	echo "Installing yt-dlp..."
108	brew install yt-dlp
109	fi
110
111	# 2. Get video title
112	VIDEO_TITLE=$(yt-dlp --print "%(title)s" "$URL" \| tr '/' '_' \| tr ':' '-' \| tr '?' '' \| tr '"' '')
113
114	# 3. Download transcript
115	yt-dlp --write-auto-sub --skip-download --sub-langs en --output "temp_transcript" "$URL"
116
117	# 4. Convert to clean text (deduplicate)
118	python3 -c "
119	import sys, re
120	seen = set()
121	vtt_file = 'temp_transcript.en.vtt'
122	try:
123	with open(vtt_file, 'r') as f:
124	for line in f:
125	line = line.strip()
126	if line and not line.startswith('WEBVTT') and not line.startswith('Kind:') and not line.startswith('Language:') and '-->' not in line:
127	clean = re.sub('<[^>]*>', '', line)
128	clean = clean.replace('&', '&').replace('>', '>').replace('<', '<')
129	if clean and clean not in seen:
130	print(clean)
131	seen.add(clean)
132	except FileNotFoundError:
133	print('Error: Could not find transcript file', file=sys.stderr)
134	sys.exit(1)
135	" > "${VIDEO_TITLE}.txt"
136
137	# 5. Cleanup
138	rm -f temp_transcript.en.vtt
139
140	CONTENT_FILE="${VIDEO_TITLE}.txt"
141	echo "✓ Saved transcript: $CONTENT_FILE"
142	```
143
144	#### Article/Blog Post
145
146	```bash
147	# Use article-extractor skill workflow
148	echo "📄 Extracting article content..."
149
150	# 1. Check for extraction tools
151	if command -v reader &> /dev/null; then
152	TOOL="reader"
153	elif command -v trafilatura &> /dev/null; then
154	TOOL="trafilatura"
155	else
156	TOOL="fallback"
157	fi
158
159	echo "Using: $TOOL"
160
161	# 2. Extract based on tool
162	case $TOOL in
163	reader)
164	reader "$URL" > temp_article.txt
165	ARTICLE_TITLE=$(head -n 1 temp_article.txt \| sed 's/^# //')
166	;;
167
168	trafilatura)
169	METADATA=$(trafilatura --URL "$URL" --json)
170	ARTICLE_TITLE=$(echo "$METADATA" \| python3 -c "import json, sys; print(json.load(sys.stdin).get('title', 'Article'))")
171	trafilatura --URL "$URL" --output-format txt --no-comments > temp_article.txt
172	;;
173
174	fallback)

People who liked this also viewed

6 picks

Type