xiaoqianran/kaggle-demo / skills
kaggle-cli
VerifiedUse the local Kaggle CLI skill for command guidance, workflows, and troubleshooting across competitions, datasets, kernels/notebooks, models, model variations and versions, inbox file uploads, forums/discussions, benchmarks, configuration, OAuth/API-token authentication, and accelerator quota. Activate this skill when the user asks about kaggle CLI commands, examples, flags, metadata files, download/upload flows, submissions, benchmark tasks, or Kaggle CLI behavior.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.2 KB |
| references/auth.md | markdown | 3.5 KB |
| references/benchmarks.md | markdown | 25.1 KB |
| references/competitions.md | markdown | 11.3 KB |
| references/configuration.md | markdown | 2.7 KB |
| references/datasets.md | markdown | 7.5 KB |
| references/files.md | markdown | 2.0 KB |
| references/forums.md | markdown | 2.9 KB |
| references/kernels.md | markdown | 6.0 KB |
| references/model_variations.md | markdown | 4.2 KB |
| references/model_variations_versions.md | markdown | 3.5 KB |
| references/models.md | markdown | 4.8 KB |
| references/quota.md | markdown | 1.2 KB |
| references/search.md | markdown | 2.4 KB |
references/kernels.md
View on GitHub# Kernels CLI Reference
Use `kaggle kernels` or alias `kaggle k` to list kernels, inspect output files,
initialize metadata, push notebook/script code, pull code, download output,
inspect run status, stream logs, and delete kernels.
## Prerequisites
- Python 3.11+ with the `kaggle` package installed.
- Kaggle credentials for push/pull/output/status/logs/delete.
- For push flows, prepare a folder containing `kernel-metadata.json`. Use
`kaggle kernels init -p <FOLDER>` to generate a starter file.
## Command Hierarchy
```text
kaggle kernels (alias: kaggle k)
├── list
├── files
├── init
├── push | update
├── pull | get
├── output
├── status
├── logs
└── delete
```
Note: `get` is an alias for `pull`; `update` is an alias for `push`.
## `kaggle kernels list`
Lists available kernels.
**Usage:**
```bash
kaggle kernels list [options]
```
**Options:**
- `-m, --mine`: Only your kernels.
- `-p, --page <PAGE>`: Page number.
- `--page-size <SIZE>`: Page size, default 20.
- `-s, --search <TERM>`: Search text.
- `--parent <KERNEL>`: Children of a parent kernel.
- `--competition <SLUG>`: Kernels for a competition.
- `--dataset <OWNER/SLUG>`: Kernels for a dataset.
- `--user <USER>`: Kernels by user.
- `--language <LANG>`: `all`, `python`, `r`, `sqlite`, `julia`.
- `--kernel-type <TYPE>`: `all`, `script`, `notebook`.
- `--output-type <TYPE>`: `all`, `visualization`, `data`.
- `--sort-by <SORT>`: Sort order.
- `-v, --csv`: Print CSV.
**Examples:**
```bash
kaggle kernels list
kaggle kernels list --user kaggle --language python --sort-by dateRun
kaggle k list -m -v
```
**Purpose:** Find kernel handles in `<owner>/<kernel-slug>` format.
## `kaggle kernels files`
Lists output files for a kernel.
**Usage:**
```bash
kaggle kernels files [KERNEL] [options]
```
**Options:**
- `--page-size <SIZE>`: Page size.
- `--page-token <TOKEN>`: Page token.
- `-v, --csv`: Print CSV.
**Examples:**
```bash
kaggle kernels files kerneler/sqlite-global-default -v --page-size 1
```
**Purpose:** Inspect generated output before downloading.
## `kaggle kernels init`
Creates a starter `kernel-metadata.json`.
**Usage:**
```bash
kaggle kernels init [options]
```
**Options:**
- `-p, --path <FOLDER>`: Folder where metadata will be written.
**Examples:**
```bash
kaggle kernels init -p my-kernel
```
**Purpose:** Bootstrap local metadata for `kaggle kernels push`.
## `kaggle kernels push`
Pushes code to a kernel and runs it.
**Usage:**
```bash
kaggle kernels push [options]
```
**Options:**
- `-p, --path <FOLDER>`: Folder containing files and `kernel-metadata.json`.
- `-t, --timeout <SECONDS>`: Limit run time, bounded by Kaggle's maximum.
- `--accelerator <ACCELERATOR>`: Accelerator type for the kernel run.
**Examples:**
```bash
kaggle kernels push -p my-kernel --timeout 3600 --accelerator gpu
kaggle kernels update -p my-kernel
```
**Purpose:** Upload local notebook/script code and create a new kernel version.
**Notes:** `NvidiaTeslaP100` is not usable for GPU compute with the default
Kaggle image, whose PyTorch build (cu128) omits Pascal (`sm_60`) kernels:
`torch.cuda.is_available()` returns `True`, but the first CUDA operation fails
with `cudaErrorNoKernelImageForDevice`. Use `NvidiaTeslaT4` or
install a Pascal-compatible torch build.
## `kaggle kernels pull`
Pulls code from a kernel.
**Usage:**
```bash
kaggle kernels pull [KERNEL] [options]
```
**Options:**
- `-p, --path <PATH>`: Download folder.
- `-w, --wp`: Download to current working path.
- `-m, --metadata`: Generate metadata when pulling.
**Examples:**
```bash
kaggle kernels pull owner/kernel-slug -p pulled
kaggle kernels get owner/kernel-slug -w -m
kaggle k pull owner/kernel-slug/3 -w -m
```
**Purpose:** Retrieve notebook/script source from Kaggle.
**Notes:** Kernel references may include an optional version:
`<owner>/<kernel-name>/<version>`.
## `kaggle kernels output`
Downloads output from the latest kernel run.
**Usage:**
```bash
kaggle kernels output [KERNEL] [options]
```
**Options:**
- `-p, --path <PATH>`: Download folder.
- `-w, --wp`: Download to current working path.
- `-o, --force`: Force download.
- `-q, --quiet`: Suppress progress output.
- `--file-pattern <REGEX>`: Download only matching output files.
- `--page-size <SIZE>`: Output files per page.
- `--page-token <TOKEN>`: Download from one output page.
**Examples:**
```bash
kaggle kernels output owner/kernel-slug -p output
kaggle k output owner/kernel-slug --file-pattern ".*\\.png$"
```
**Purpose:** Retrieve generated files such as submissions, images, or processed
data.
**Notes:** Without `--page-token`, output download scans available pages so
`--file-pattern` can match files beyond the first page.
## `kaggle kernels status`
Displays status of the latest kernel run.
**Usage:**
```bash
kaggle kernels status [KERNEL]
```
**Options:**
- No visible options. If `[KERNEL]` is omitted, local `kernel-metadata.json`
supplies the kernel reference.
**Examples:**
```bash
kaggle kernels status owner/kernel-slug
```
**Purpose:** Check whether the latest run is queued, running, complete, or
errored.
## `kaggle kernels logs`
Prints execution logs from the latest kernel run.
**Usage:**
```bash
kaggle kernels logs [KERNEL] [options]
```
**Options:**
- `-f, --follow`: Continuously poll and print new log lines.
- `--interval <SECONDS>`: Poll interval for follow mode. Default 5.
**Examples:**
```bash
kaggle kernels logs owner/kernel-slug
kaggle k logs owner/kernel-slug --follow --interval 10
```
**Purpose:** Debug kernel execution from CLI logs.
## `kaggle kernels delete`
Deletes a kernel.
**Usage:**
```bash
kaggle kernels delete <KERNEL> [options]
```
**Options:**
- `-y, --yes`: Skip confirmation.
**Examples:**
```bash
kaggle kernels delete owner/kernel-slug -y
```
**Purpose:** Remove a kernel from Kaggle.
## Error Scenarios And Notes
- Kernel handles are `<owner>/<kernel-name>` or `<owner>/<kernel-name>/<version>`.
- Push requires a metadata file and valid code file path in that metadata.
- `logs --follow` polls repeatedly; use `--interval` to avoid excessive calls.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |
| agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |