Skip to content
Get packs

Guide · Agent skill

How to use the BigQuery Basics agent skill

Updated · written with AI from the project's own documentation and checked against our catalogue data

The BigQuery Basics skill gives an AI agent working knowledge of Google BigQuery, a cloud data warehouse that runs SQL over very large datasets. This guide explains what an agent skill is, how to install this one, how to ask for work in plain language, and what to check before letting it touch a real Google Cloud project. By the end you will be able to set up a dataset, create a table and run a first query through your agent.

BigQuery Basicsgoogle/skills · Apache-2.0

Quick start

  1. 1Download and unzip the skill. Download the ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder.
  2. 2Or upload it in the Claude apps. Go to Settings → Capabilities → Skills and upload the ZIP file.
  3. 3Prepare Google Cloud access. Install the gcloud SDK and bq tool, and confirm your project has the BigQuery API enabled and permission to create resources.
  4. 4Make a natural request. Describe the outcome, such as creating a dataset or running a query, and include project, dataset and table names.
  5. 5Review the commands. Check any command that enables APIs or creates resources before allowing it to run.
  6. 6Run a test query. Start with the public usa_names query to confirm the setup works.

What the BigQuery Basics skill makes the agent do

An agent skill is a folder of instructions and reference files that an AI agent loads when a task matches it. The entry file, SKILL.md, tells the agent when to use the skill and how to approach the work. Here, the skill covers managing datasets, tables and jobs in BigQuery, running SQL queries, and basic data ingestion and analysis.

Once installed, the agent can create datasets and tables, run queries, and point you to the right commands, client library code, or Terraform configuration. For more detailed topics it reads reference files covering core concepts, change history, continuous queries, the bq command-line tool, client libraries, MCP usage, infrastructure as code, IAM and security, and usage attribution.

  • Create datasets and tables, including from a schema file
  • Run SQL queries against BigQuery
  • Explain change history, continuous queries, IAM roles and attribution tags
  • Provide client library code for Python, Java, Node.js or Go, or Terraform config

Install the BigQuery Basics skill

Download the skill as a ZIP file from this page and unzip it. You can then place the folder in one of two locations depending on how you use your agent.

For a personal setup available across projects, unzip it into your user skills folder at ~/.claude/skills/. To limit it to one project, unzip it into the .claude/skills/ folder inside that project. In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.

  • Personal (all projects): ~/.claude/skills/
  • Single project: .claude/skills/ inside your project folder
  • Claude apps: Settings → Capabilities → Skills, then upload the ZIP

Set up Google Cloud before you start

The skill runs commands, so the tools and access it needs must be in place first. It expects the Google Cloud SDK (gcloud) and the bq command-line tool, a Google Cloud project with the BigQuery API enabled, and billing or credentials that allow creating resources. If you plan to use code examples, you also need the Google Cloud client libraries for your language.

Optional extras are the BigQuery remote MCP server or the Gemini CLI extension, and the Developer Knowledge MCP server for extra product lookups.

  • Google Cloud SDK (gcloud) and bq command-line tool
  • A Google Cloud project with the BigQuery API enabled
  • Billing or credentials that can create resources
  • Client libraries for Python, Java, Node.js or Go (only for code examples)
  • Optional: BigQuery remote MCP server or Gemini CLI extension
  • Optional: Developer Knowledge MCP server

Trigger the skill with natural requests

You do not need to name the skill. The agent loads it when your request matches its description, such as wanting to run SQL, manage BigQuery datasets, tables or views, or ingest and analyse data. Describe the outcome you want and include the names you care about.

Be specific about the project, dataset and table names. The agent will then choose the matching commands from the reference files.

  • "Create a BigQuery dataset called sales_demo in the US location."
  • "Make a table in sales_demo from my schema.json file."
  • "Run a query that returns the ten most common names in the public USA names dataset for Texas."

Run the basic workflow

The skill's setup covers four basic steps: enable the BigQuery API, create a dataset, create a table from a JSON schema file, and run a query. The example below uses the public usa_names dataset, so you can test a query without loading your own data.

Enabling the API is done with gcloud, and the skill tags that command with attribution metadata as described in its attribution notes. Review commands before they run, especially any that enable APIs or create resources.

bq mk --dataset --location=US my_dataset bq mk --table my_dataset.mytable schema.json bq query --use_legacy_sql=false 'SELECT name FROM `bigquery-public-data.usa_names.usa_1910_2013` WHERE state = "TX" LIMIT 10'

Safety notes: scripts and network access

This skill runs scripts and uses the network. It can run commands that enable APIs and create resources, so a mistake can cost money or expose data. Treat production changes to billing, IAM or shared datasets as work that needs human review before the agent runs it.

The attribution notes ask the agent to tag gcloud and HTTP calls with a label that identifies only the skill and its version. The skill says the tag carries no user data and avoids gcloud config set so the tag does not persist beyond the task.

  • Review any command that enables APIs, creates resources or changes IAM before approving it
  • Use a test project for first runs
  • Do not use this skill for production billing or shared datasets without review

Tips for getting good results

Give the agent the exact table schema or a schema file, since table creation depends on field names, types and modes. For change tracking, note that the CHANGES function needs the table's enable_change_history option set to TRUE, while APPENDS does not. For permissions questions, ask the agent to check the IAM and security reference rather than guessing roles.

  • Provide schema.json with field names, types and modes
  • Ask for the bq command first, then run it yourself
  • Mention the location (for example US) when creating datasets

What people use it for

  • Creating a sales dataset and table from a schema file for a first analysis
  • Running SQL queries over a public dataset to learn BigQuery
  • Tracking incremental table changes with the CHANGES or APPENDS functions
  • Generating Terraform config for datasets and tables in a pipeline
  • Getting bq CLI commands explained without memorising the syntax

Frequently asked questions

What is the BigQuery Basics agent skill?

It is an instruction folder for an AI agent that gives it working knowledge of Google BigQuery. It covers managing datasets, tables and jobs, running SQL, and pointing you to the right commands, client library code or Terraform config.

What do I need to use the BigQuery skill?

You need the Google Cloud SDK (gcloud) and the bq command-line tool, a Google Cloud project with the BigQuery API enabled, and billing or credentials that let you create resources. Client libraries are only needed for code examples.

Is the BigQuery Basics skill free?

The skill is licensed under Apache-2.0. Google Cloud usage, such as running queries or creating resources, may incur charges on your project, and this page does not state the pricing.

Where do I put the skill so Claude can find it?

Unzip it into ~/.claude/skills/ for personal use across projects, or into .claude/skills/ inside one project. In the Claude apps you can upload the ZIP under Settings → Capabilities → Skills.

Does the skill run commands on my computer?

Yes. The skill runs scripts and uses the network, and it can run commands that enable APIs and create resources. Review commands before approving them, and avoid using it on production billing, IAM or shared datasets without review.

What are the alternatives if I do not want an agent to run BigQuery commands?

You can run the bq command-line tool or the BigQuery console yourself, or use Terraform or client libraries directly. The skill mainly helps when you want an agent to write and run these steps for you.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

Use this skill if you are a data analyst, engineer or developer who wants an agent to run BigQuery SQL, create datasets and tables, or generate Terraform and client code, and you have a Google Cloud project to test in. Skip it if you need general SQL help without a Google Cloud account, or if you want to make unsupervised changes to billing, IAM or shared production data.

Download the skill