kaggle-competitions

List all competitions

$ kaggle [[c|competitions]] list

$ kaggle [[c|competitions]] download [competition_name]

$ kaggle [[c|competitions]] download [competition_name] [[-f|--file]] [file]

$ kaggle [[c|competitions]] submit [competition_name] [[-f|--file]] [path/to/file] [[-m|--message]] "[message]"

$ kaggle [[c|competitions]] leaderboard [competition_name] [--show|--download]

$ kaggle [[c|competitions]] submissions [competition_name]

kaggle competitions <action> [competition] [options]

list
    List competitions matching criteria

submit
    Submit a file to a specific competition

download
    Download competition dataset

files
    List files in a competition

leaderboard
    Download competition leaderboard

leaderboards
    List leaderboards across competitions

overview
    Print competition overview page

-h, --help
    Show help for the command or subcommand

--page-size PAGE_SIZE
    Number of items to retrieve (default 20, max 5000)

--category CATEGORY
    Filter by category (featured, research, recruiting, playground, gettingStarted)

--organization ORG
    Filter by organization team

--status STATUS
    Filter by status (active, completed, featuredActive, featuredCompleted)

--sort-by SORT_BY
    Sort by (prize, submissions, entries, daysActive)

-c COMPETITION, --competition COMPETITION
    Competition slug or ID

-f FILE, --file FILE
    Path to submission file

-m MESSAGE, --message MESSAGE
    Optional message for the submission

--dir DIR
    Download data to specific directory

-q, --quiet
    Suppress output

--retry RETRY_COUNT
    Number of retries for submission

--team TEAM_ID
    Submit on behalf of team

--public
    Download public leaderboard (for leaderboard)

--gpu GPU
    Filter by GPU usage (T4 x2, P100, etc.)

The kaggle competitions command is a subcommand of the official Kaggle CLI tool, enabling seamless interaction with Kaggle's data science competitions from the Linux terminal. Users can list active competitions, submit prediction files, download datasets and sample submissions, inspect files, view leaderboards, and more.

Installation requires Python's pip: pip install kaggle. Authentication involves downloading kaggle.json from your Kaggle account settings and placing it in ~/.kaggle/kaggle.json with permissions chmod 600.

Example workflows include discovering competitions with kaggle competitions list --status Active, preparing submissions via kaggle competitions download -c titanic -p /path/to/dir, and uploading results using kaggle competitions submit -c titanic -f submission.csv -m "Initial model". This tool boosts productivity by automating repetitive tasks, integrating with scripts, and supporting bulk operations for competitive data science.

Requires valid Kaggle API credentials and internet; subject to API rate limits (e.g., 20 calls/min); large downloads may take time; submissions must match competition format or fail.

pip install kaggle
Download kaggle.json from Kaggle > Account > API.
mkdir ~/.kaggle && mv kaggle.json ~/.kaggle/ && chmod 600 ~/.kaggle/kaggle.json

Run kaggle competitions list; success confirms valid API key.

Introduced in 2018 as part of Kaggle's official Python API (v1.5+), evolving from community scripts to support growing competition ecosystem; major updates in 2020-2023 added leaderboard downloads, team submissions, and GPU filtering amid rising competition participation.

kaggle(1), pip(1), curl(1)