bigquery-basics Skill (Google) for Claude Code: BigQuery

Lets your agent manage Google BigQuery datasets, tables and jobs and run SQL queries with the bq and gcloud tools.

Installs: 18.3K; on skills.sh
Stars: 21.8K; on GitHub
Files: 9; besides SKILL.md
Updated: Nov 2026; last repo commit

The skill gives your agent a short BigQuery workflow. It enables the BigQuery API with gcloud, creates a dataset and a table from a schema file with bq, and runs standard SQL queries. Nine reference pages add core concepts, the bq command guide, client libraries, Terraform, IAM and security, change history, continuous queries and the BigQuery MCP server.

It needs the Google Cloud tools and a project you can use. Queries cost money based on the data they scan, so ask the agent to do a dry run first. The skill also tells the agent to tag each gcloud command with a usage attribution label, which you can switch off by asking.

Good for

Analysts who want an agent to write and run BigQuery SQL
Cloud engineers creating datasets, tables and views from the terminal
Developers wiring BigQuery client libraries or Terraform

Install bigquery-basics

Claude Code

Copy this prompt.

Install the bigquery-basics skill in Claude Code. Run this command:

npx skills add https://github.com/google/skills --skill bigquery-basics

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

From its listing on skills.sh. Source.

What's in the skill

When it loads

Manages datasets, tables, and jobs in BigQuery. Use when you need to interact with BigQuery, run SQL queries, manage BigQuery resources (datasets, tables, views), or perform basic data ingestion and analysis.

Sections in SKILL.md

Attribution
Setup and Basic Usage
Reference Directory
Related Skills

Key facts

Author: Google (google/skills)
Repo: google/skills, folder skills/cloud/bigquery-basics
Installs on skills.sh: 18,334 as of 2026-11-02
GitHub stars: 21,764 for the whole google/skills repo
License: Apache-2.0 (as detected by GitHub)
Last repo commit: 2026-11-01
What is inside: 10 files including SKILL.md; 9 reference pages (concepts, CLI, client libraries, MCP, Terraform, IAM, change history, continuous queries, attribution)
skills.sh audits: Gen Agent Trust Hub: pass; Socket: pass; Snyk: warn
Version: 1.0.2 (SKILL.md frontmatter)

What it includes

gcloud services enable bigquery.googleapis.com
Turns on the BigQuery API for your Google Cloud project. Runs with an attribution label set.
Runs code · Enable the BigQuery API
Create a dataset
Creates a dataset in a chosen location, US in the example.
Runs code · bq mk --dataset
Create a table from a schema file
The agent writes a schema.json, then creates the table from it.
Runs code · bq mk --table
Run standard SQL
Runs a query with legacy SQL off. The CLI guide also shows a dry run that estimates bytes processed before you spend anything.
Try: “Dry-run this query against my dataset and tell me how many bytes it would process.”
Runs code · bq query
Copy, load, insert, delete, list, cancel
The CLI usage page covers copying tables and snapshots, loading from Cloud Storage, inserting rows, deleting a table, and listing, showing and cancelling jobs.
Delete · Manage tables and jobs
Python, Java, Node.js, Go and IaC
Reference pages show client library setup, including a Python pip install, and Terraform examples for datasets, tables and reservations.
Write · Client libraries and Terraform
Least-privilege roles and data controls
Advice on granting the least permissive role, service account impersonation, VPC Service Controls and column and row security.
Read · IAM and security
The BigQuery remote MCP server
Describes the tools of the BigQuery MCP server. Its SQL tool is limited to SELECT statements.
Read · MCP usage
Label gcloud and API calls with the skill
Tells the agent to set an attribution label on each gcloud command and add a User-Agent suffix on direct HTTP calls. It carries no user data, and you can opt out.
Runs code · Usage attribution

From the official docs

Permissions

Medium risk

It runs gcloud and bq against your real Google Cloud project, so it can create and delete tables and run billed queries. Use a test project or a limited role, and ask for a dry run before large queries.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It can create and delete tables
The workflow creates datasets and tables, and the CLI guide shows bq rm -f for deleting a table. Use a project you can afford to change and approve each command. github.com
Queries are billed by data scanned
The core concepts page says analysis is billed by bytes processed on demand, or by slots. Ask for a dry run first, as the CLI guide shows. github.com
gcloud commands carry an attribution tag
The agent sets an environment variable and a User-Agent suffix naming this skill and its version. Ask it to turn this off if you do not want the tag. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	No read-only switch in the skill. Use a Google Cloud account or role limited to read access if you only want queries.
Talks to	Google Cloud receives your BigQuery commands and queries. Each gcloud command is tagged with an attribution label naming the skill. pip install --upgrade google-cloud-bigquery appears in the client library guide
Sign-in and keys	Uses whatever Google Cloud login your gcloud and bq tools already have.
Scopes	Runs gcloud and bq in your terminal with your Google Cloud login Creates datasets and tables, and the CLI guide includes deleting a table Enables the BigQuery API on your project

Questions

What does the bigquery-basics skill do?

It makes your agent work with Google BigQuery. It enables the API, creates datasets and tables, runs SQL queries and manages jobs using the gcloud and bq tools.

How do I install it?

Use the install box at the top of this page for this one skill. The google/skills README also gives a command for the whole collection and plugin routes for Claude Code and Codex.

Does it need an API key?

No key in the skill itself. It uses your Google Cloud login in gcloud and bq, and your account needs permission on the project.

Will running it cost money?

Queries can. The reference says analysis is billed by bytes processed on demand, and storage is billed too. The CLI guide shows a dry run that estimates bytes before you run a query.

Can it delete my data?

The CLI guide includes a command that deletes a table, and the agent runs commands with your login. Approve each command, or use a role that cannot delete.

What is the attribution tag?

The skill has the agent set a metrics label on each gcloud command and add a User-Agent suffix on direct HTTP calls. It names the skill and version and carries no user data. If you ask to disable it, the skill says to omit it.

Does it cover BigQuery AI and ML?

No. It points to a separate bigquery-ai-ml skill for forecasting, anomaly detection and text generation.

Information

updated Nov 2, 2026

Repository: google/skills
Folder: skills/cloud/bigquery-basics

google/skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 13

bigquery-basics SKILL.md · github.com
google/skills README · github.com
bigquery-basics attribution.md · github.com
bigquery-basics cli-usage.md · github.com
bigquery-basics core-concepts.md · github.com
bigquery-basics mcp-usage.md · github.com
bigquery-basics client-library-usage.md · github.com
bigquery-basics iac-usage.md · github.com
bigquery-basics iam-security.md · github.com
bigquery-basics change-history.md · github.com
bigquery-basics continuous-queries.md · github.com
bigquery-basics on skills.sh · skills.sh
google/skills · github.com