Skip to main content
notewise process is the main command. It fetches a YouTube transcript, sends it to your configured LLM, and writes structured Markdown study notes to disk. 
notewise process URL [OPTIONS]

​
What URL accepts

The URL argument accepts three input types:
  • Single video URL — https://youtube.com/watch?v=VIDEO_ID
  • Playlist URL — https://youtube.com/playlist?list=PLAYLIST_ID
  • Batch file path — path to a .txt file containing one URL per line (e.g. urls.txt)
For batch files, notewise automatically detects whether the argument is a URL or a file path — no extra flag needed.

​
Options

​
--model
string
LLM model string, in LiteLLM format. Overrides the DEFAULT_MODEL setting in your config file.Default: gemini/gemini-2.5-flashAliases: -mExamples: gpt-4o, claude-3-5-sonnet-20241022, groq/llama3-70b-8192
​
--output
path
Directory where study notes are written. Created automatically if it does not exist. Overrides the OUTPUT_DIR config setting.Default: ./outputAliases: -o
​
--language
string
Preferred transcript language code (e.g. en, hi, es). Repeat the flag to supply a priority-ordered fallback list. Overrides the default_languages config setting.Default: enAliases: -lThis flag is repeatable — pass it multiple times to specify fallback languages in order:
notewise process "URL" --language hi --language en
​
--temperature
number
LLM sampling temperature. Controls how deterministic the generated notes are. Must be between 0.0 (most deterministic) and 1.0 (most creative). Overrides the TEMPERATURE config setting.Default: 0.7Aliases: -t
​
--max-tokens
integer
Maximum number of tokens per LLM response. Useful for controlling cost on long videos or when using models with tight context windows. Must be at least 1. Overrides the MAX_TOKENS config setting.Default: model default (no limit imposed by notewise)Aliases: -k
​
--force
boolean
Re-process a video even if its output already exists in the output directory and the local cache. Without this flag, already-processed videos are skipped automatically.Default: falseAliases: -F
​
--no-ui
boolean
Disable the Rich live dashboard and output plain progress lines to stdout instead. Recommended for CI pipelines, cron jobs, and any environment where a live terminal UI is unavailable.Default: false
​
--quiz
boolean
Generate a multiple-choice quiz file alongside the study notes for each processed video.Default: false
​
--export-transcript
string
Export the raw transcript to a file in the video’s output directory. Accepted values:
  • txt — plain text transcript
  • json — timestamped JSON transcript
Default: not exported
​
--cookie-file
path
Path to a Netscape-format cookies .txt file. Required for processing age-gated, members-only, or other login-restricted videos. Overrides the YOUTUBE_COOKIE_FILE config setting.Default: noneAliases: --cookies

​
Examples

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

​
Output structure

Study notes are written under the output directory, organized by video title: 
output/
├── Video Title/
│   ├── study_notes.md
│   ├── quiz.md               # only with --quiz
│   └── transcript.txt        # only with --export-transcript txt
└── Another Video Title/
    └── study_notes.md
For videos longer than one hour that include YouTube chapters, notewise generates a separate notes file per chapter: 
output/
└── Long Course Title/
    ├── Chapter 01 - Introduction.md
    ├── Chapter 02 - Core Concepts.md
    └── Chapter 03 - Practical Examples.md

​
Exit codes

CodeMeaning
0All videos processed successfully
1One or more failures, or the run was interrupted
When processing a playlist or batch file, notewise continues on per-video failures and returns exit code 1 at the end if any video failed. Check the session log for individual failure details: notewise logs --tail 50.

Build docs developers (and LLMs) love

Get started for freeTalk to us