Skip to main content
Generate comprehensive, structured Markdown study notes from YouTube videos or playlists.

Synopsis

notewise process URL [OPTIONS]

Arguments

URL
string
required
YouTube video URL, playlist URL, bare video/playlist ID, or path to a .txt batch file containing one URL per line.

Options

--model
string
Alias: -mLiteLLM model string. Overrides DEFAULT_MODEL from config for this run.
notewise process "URL" --model gpt-4o
notewise process "URL" -m anthropic/claude-3-5-sonnet-20241022
--output
path
Alias: -oOutput directory path. Overrides OUTPUT_DIR from config for this run.
notewise process "URL" --output ~/my-notes
--language
string
Alias: -l · RepeatablePreferred transcript language code. Repeat the flag to set a fallback chain. Default: en.
notewise process "URL" --language hi       # Hindi first
notewise process "URL" -l fr -l en         # French, then English
--temperature
float
default:"0.7"
Alias: -tLLM sampling temperature, range 0.01.0. Lower = more deterministic. Higher = more creative.
notewise process "URL" --temperature 0.3
--max-tokens
integer
Alias: -kMaximum tokens per LLM response. When not set, the model’s own default is used.
notewise process "URL" --max-tokens 2000
--quiz
boolean
default:"false"
Also generate a multiple-choice quiz (<title>_quiz.md) alongside the notes. Contains 10–15 questions with answers and explanations.
notewise process "URL" --quiz
--export-transcript
string
Export the raw transcript alongside the notes. Accepted values: txt (plain text) or json (timestamped array).
notewise process "URL" --export-transcript txt
notewise process "URL" --export-transcript json
--force
boolean
default:"false"
Alias: -FRe-process videos even if the output directory already exists and the video is in the cache.
notewise process "URL" --force
--no-ui
boolean
default:"false"
Disable the Rich live dashboard. Prints plain progress lines to stdout — suitable for cron jobs, CI pipelines, and log aggregators.
notewise process "URL" --no-ui
Alias: --cookiesPath to a Netscape-format .txt cookies file. Required for private, age-gated, or members-only videos.
notewise process "URL" --cookie-file ./youtube-cookies.txt

Input types

notewise process "https://youtube.com/watch?v=dQw4w9WgXcQ"
notewise process "https://youtu.be/dQw4w9WgXcQ"
notewise process "https://youtube.com/shorts/dQw4w9WgXcQ"
notewise process dQw4w9WgXcQ      # bare 11-character video ID

Output structure

output/
├── Video Title.md
├── Video Title_quiz.md              # only with --quiz
├── Video Title_transcript.txt       # only with --export-transcript txt
└── Video Title_transcript.json      # only with --export-transcript json

Exit codes

CodeMeaning
0All videos processed successfully
1One or more videos failed, or an unexpected error occurred

Examples

# Basic
notewise process "https://youtube.com/watch?v=VIDEO_ID"

# Custom model and output directory
notewise process "URL" --model gpt-4o --output ~/notes

# Playlist with quiz and transcript export
notewise process "https://youtube.com/playlist?list=PLxxxxxx" \
  --quiz --export-transcript txt

# Batch file — non-interactive with high concurrency
MAX_CONCURRENT_VIDEOS=10 notewise process urls.txt --no-ui

# Private video with cookie file
notewise process "URL" --cookie-file ./youtube-cookies.txt

# Re-process with a different model
notewise process "URL" --model claude-3-5-sonnet-20241022 --force

# Preferred language fallback chain
notewise process "URL" -l hi -l en