Kaggle MCP integration
Searches datasets, models, and competitions, publishes and versions datasets, pulls notebook code and output, and files competition entries.
35actions available
Three actions you can hand over today
Every action runs live through MCP. Nothing to build, nothing to maintain.
Download kaggle dataset
Lyro accesses public datasets for analysis and comparison. Your team enriches support answers with real data and benchmarks instead of opinions.
Download competition leaderboard
Lyro retrieves performance rankings for context. Your team sees how solutions rank and recommends top-performing approaches to customers.
Pull kernel code
Lyro fetches analysis scripts and notebooks for reference. Your team learns proven techniques and adapts them to solve customer-specific problems.
How businesses use Kaggle + Lyro
Each card is one request a support team gets, and the Kaggle actions Lyro runs to close it.
Find a dataset or model worth starting from
Lyro searches the public catalogue of datasets, models, and open competitions and reads a candidate's metadata, so the shortlist is settled before anyone downloads a multi-gigabyte archive.
List Kaggle DatasetsList Kaggle ModelsGet Dataset MetadataPublish and version a dataset of your own
Lyro writes the metadata skeleton, publishes the first version, and pushes each later revision as a new one, so a dataset the team maintains has a history rather than a single overwritten upload.
Kaggle Dataset InitDataset CreateCreate Dataset VersionPull a public notebook's code and results
Lyro retrieves a kernel's source, checks whether its last run finished, and downloads the output files, so a published approach can be read and reused instead of reverse-engineered from a screenshot.
Pull Kernel CodeGet Kernel StatusDownload Kernel OutputEnter a competition and track the result
Lyro downloads the competition's data files, files the submission, and reads the leaderboard afterwards, so an entry and its standing are handled without leaving the conversation.
Download Competition Data FilesSubmit Competition EntryView Competition Leaderboard
How it works
Get started in 3 steps
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Lyro reads the Kaggle actions it has and picks the ones a request needs.
- 01
Connect Kaggle
Authorize the Kaggle account your team already uses — one consent screen, no API keys, no mapping tables. Lyro can only do what you granted that account, and you can disconnect it at any time.
- 02
Tell your agent what you need
Describe the job the way you would hand it to a teammate. Lyro maps it to the Kaggle actions that close it and chains as many as the request needs.
- 03
Watch it work
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started free
Everything else about Kaggle
Setup, permissions, and the limits of what Lyro can do inside Kaggle.
With the same API credentials the Kaggle CLI uses, and the toolkit exposes that configuration directly. Initialize Kaggle Configuration, Set Kaggle Configuration, View Kaggle Configuration, and Reset Kaggle Configuration manage the stored keys and defaults, so a wrong username or an expired token is diagnosed by reading the config rather than by guessing at a 403.
Every action available in Kaggle
All 35 actions your agent can call on Kaggle, straight from the live MCP connection.
Download competition data files
Downloads all data files for a Kaggle competition as a single zip archive.
Submit competition entry
Submit an entry to a Kaggle competition using a previously uploaded file.
Get kaggle config directory
Retrieve the directory of the Kaggle API configuration file.
Initialize kaggle configuration
Initialize Kaggle API client configuration.
List kaggle configuration keys
List local Kaggle API configuration keys.
Get kaggle config path
Retrieve local Kaggle API configuration file path.
Reset kaggle configuration
Reset local Kaggle CLI configuration to defaults.
Set kaggle configuration
Set a Kaggle CLI configuration parameter.
Unset kaggle configuration
Unset a Kaggle CLI configuration parameter.
View kaggle configuration
View local Kaggle API credentials and configuration settings.
Dataset create
Create a new Kaggle dataset with metadata.
Kaggle dataset init
Initialize a dataset-metadata.json file in a local folder.
List kaggle dataset files
List files in a Kaggle dataset.
Get dataset status
Check the processing status of a Kaggle dataset after creation or version update.
Create dataset version
Create a new version of an existing Kaggle dataset.
Download competition file
Download a specific data file from a Kaggle competition.
Download competition leaderboard
Download the entire competition leaderboard as a CSV file packaged in a ZIP archive.
Download kaggle dataset
Download all files from a Kaggle dataset as a zip archive.
Download kaggle dataset file
Download a specific file from a Kaggle dataset.
Generate competition submission URL
Generate a pre-signed URL for uploading competition submission files.
Get dataset metadata
Get comprehensive metadata for a Kaggle dataset including title, description, licenses, and tags.
Get model details
Get a Kaggle model's details including metadata and description.
Get model instance details
Get details for a specific Kaggle model instance (variation).
Kaggle kernel init
Initialize a kernel-metadata.json template file in a specified folder.
Download kernel output
Download the output of a Kaggle kernel.
Get kernel status
Get the execution status of a Kaggle kernel (notebook).
List competition data files
List all data files available for a Kaggle competition.
List kaggle competitions
List available Kaggle competitions with filters and pagination.
List kaggle datasets
List Kaggle datasets with filters and pagination.
List kernel output files
List output files for a specific kernel run.
The tools Kaggle sits next to
Same connection, same setup. Pick the next one your team already uses.
Chatbotkit
Pulls bot transcripts and usage figures, clones blueprints into new bots, resyncs training datasets, and wires up messaging channels.
Flowiseai
Clones and updates chatflows, rewrites document store chunks behind bad answers, reads collected leads and feedback, and health-checks the instance.

Linkup
Searches the live web with date and domain filters, returns sourced answers or structured JSON, and fetches a page as markdown.
Pinecone
Runs similarity searches and reranks results, upserts records with generated embeddings, provisions indexes and namespaces, and runs backups and imports.
Studio By Ai21 Labs
Creates and edits AI21 assistants, validates the Python behind their plans, indexes site content, and runs an assistant on request.
Zep
Writes thread messages into a user graph, retrieves the context relevant to the conversation, searches it semantically, and deletes data on request.

Ready to connect Kaggle?
Authorize the account and your agent has all 35 actions from the first conversation.


