Kaggle CLI
For a repeatable experiment, an agent downloads a Kaggle dataset to a working directory and retrieves notebook outputs.
- 7,644 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- The client uses Kaggle account permissions, downloads data and can publish, update or delete account resources.
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/kaggle-cliDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The client uses Kaggle account permissions, downloads data and can publish, update or delete account resources.
This is third-party code. Review the repository files before installing.
What it does
Kaggle CLI finds competitions and datasets and downloads data files, notebook code and outputs. An agent performs these steps from the terminal while the user records resource identifiers for the next experiment. The same client can publish or delete datasets and models, run notebooks and submit competition entries. Commands use the signed-in Kaggle user’s permissions. A separate command group browses discussion forums.
Who it is for. Kaggle participants reproducing data downloads and experiment preparation with an agent.
Good fit when
- An experiment needs files from a particular Kaggle dataset.
- A user retrieves code and output after a notebook run.
- A new version of an owned dataset is ready for upload.
- A team prepares a submission to a selected competition.
Not a fit when
- Data access or usage terms have not been agreed.
- Work products must not be published or submitted from the account.
- You need a full local notebook editor rather than command access.
Example request
Find the approved dataset and prepare its download to a dedicated directory without creating or modifying account resources.Limitations
The current CLI requires Python 3.11 or later. On Linux with a protected system Python, pip installation may fail; `pipx install kaggle` creates an isolated environment. Account operations require Kaggle authentication, and file access depends on the resource’s permissions and terms. Creation, update and deletion affect real account data, so the user must select the exact dataset, model or notebook before running them.
How to disable. Stop Kaggle jobs and remove unused stored credentials. Uploaded datasets, models and submissions remain in the account until separately handled.
FAQ
Where are the command groups listed?
Run `kaggle --help`.
How do I start browser authentication?
Use `kaggle auth login`.
How do I start looking for a dataset?
Run `kaggle datasets list` to obtain a list, then copy the required dataset identifier for subsequent commands.
Why is kaggle missing from the shell after installation?
Add the Python scripts directory to `PATH`. A user install on Linux usually uses `~/.local/bin`; on Windows use the `Scripts` directory of your Python installation.
Can I read discussions?
Yes. The CLI browses forums and reads threads.
Discussion
No questions or reviews yet.
Related
CLI toolsA skill that runs a Kaggle competition from rules and folds to a scored submission, with hidden-rerun debugging and ensembles
SearchCLI is a Volcengine AI Search tool that evaluates results against test queries and retains an experiment plan and report.
Official ScrapeAtlas terminal client: an agent or script pulls public data from 49 platforms with one command and gets plain JSON
The CLI imports CSV and JSON into SQLite, runs SQL and reshapes tables for agent-assisted local data preparation.
Official Databricks skills guide coding agents through data work, Lakeflow jobs and resource deployment on the platform.
A free open-source app where the AI assistant Kady handles research work with data and literature on your computer
README in short
Authentication supports browser login, tokens and the legacy kaggle.json format. CLI development instructions are separate from ordinary installation.