Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Bigquery Ml Model Creator · dicklesworthstone | DecimalAI

dicklesworthstone/bigquery-ml-model-creator

SkillScore100Maintained

Create bigquery ml model creator operations. Auto-activating skill for GCP Skills. Triggers on: bigquery ml model creator, bigquery ml model creator Part of the GCP Skills skill category. Use when working with bigquery ml model creator functionality. Trigger with phrases like "bigquery ml model creator", "bigquery creator", "bigquery".

by dicklesworthstone

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

AI Engineer pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/dicklesworthstone-bigquery-ml-model-creator/SKILL.md

decimalai skills pull dicklesworthstone-bigquery-ml-model-creator --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/dicklesworthstone-bigquery-ml-model-creator/SKILL.md https://app.decimal.ai/s/dicklesworthstone-bigquery-ml-model-creator/SKILL.md

Raw file: /s/dicklesworthstone-bigquery-ml-model-creator/SKILL.md ↗ · machine-readable summary at /s/dicklesworthstone-bigquery-ml-model-creator.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 20d ago

Yes — +14% lift+9% tokens0% turns100% of whole cases pass with skill (22/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	-16%	0%
case-02	✗→✓	▲ Improved	-8%	0%
case-17	✗→✓	▲ Improved	4%	0%
case-12	✓→✓	＝ Same ✓	11%	0%
case-03	✓→✓	＝ Same ✓	13%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

dicklesworthstone / pi_agent_rust

tests/ext_conformance/artifacts/plugins-community/skills/14-gcp-skills/bigquery-ml-model-creator/SKILL.md

View on GitHub ↗

License MITImported 8/17/2026Version v1

Skill contents

SKILL.md

Bigquery Ml Model Creator

Overview

This skill provides automated assistance for bigquery ml model creator tasks within the GCP Skills domain.

When to Use

This skill activates automatically when you:

Mention "bigquery ml model creator" in your request
Ask about bigquery ml model creator patterns or best practices
Need help with google cloud platform skills covering compute, storage, bigquery, vertex ai, and gcp-specific services.

Instructions

Provides step-by-step guidance for bigquery ml model creator
Follows industry best practices and patterns
Generates production-ready code and configurations
Validates outputs against common standards

Examples

Example: Basic Usage Request: "Help me with bigquery ml model creator" Result: Provides step-by-step guidance and generates appropriate configurations

Prerequisites

Relevant development environment configured
Access to necessary tools and services
Basic understanding of gcp skills concepts

Output

Generated configurations and code
Best practice recommendations
Validation results

Error Handling

| Error | Cause | Solution | |-------|-------|----------| | Configuration invalid | Missing required fields | Check documentation for required parameters | | Tool not found | Dependency not installed | Install required tools per prerequisites | | Permission denied | Insufficient access | Verify credentials and permissions |

Resources

Official documentation for related tools
Best practices guides
Community examples and tutorials

Related Skills

Part of the GCP Skills skill category. Tags: gcp, bigquery, vertex-ai, cloud-run, firebase

Illustrative Want to see what dicklesworthstone/bigquery-ml-model-creator would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

9/8/2026, 05:30 UTC · 22 cases

Lift

+14%

100% vs 86%

Agent turns

1.0 vs 1.0

Tokens

+9%

1,998.27 vs 1,827.09

fail → pass · 3pass → pass · 19

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-12 Construct a BigQuery SQL query to run batch predictions using model `models.customer_segment_model` on fresh feature table `staging.new_customers`.	pass→pass	12,961	11,945	-8%	1	1	0%	1,459	1,626	+11%	0	0	—
case-01 Write a SQL query in Google BigQuery to train a customer churn classifier directly on table analytics.customer_events. The target binary column is churned. I might be tempted to extract the dataset into Python scikit-learn, but write the SQL statement to perform model training directly inside BigQuery.	fail→pass	17,779	13,941	-22%	1	1	0%	2,280	1,904	-16%	0	0	—
case-02 Construct a BigQuery SQL query to predict continuous transaction amounts in dataset.sales using features user_age and historical_spend. I am considering writing a PySpark script on Dataproc, but produce the BigQuery statement to train a continuous prediction model natively.	fail→pass	15,959	13,039	-18%	1	1	0%	2,020	1,853	-8%	0	0	—
case-03 Write a BigQuery query to create a daily demand forecast model on table store.daily_sales using timestamp column sale_date and target column total_units. Instead of building a simple linear regression over time, train an automated ARIMA time series model.	pass→pass	13,488	12,290	-9%	1	1	0%	1,544	1,745	+13%	0	0	—
case-04 Create a BigQuery query to group user profiles from table analytics.user_features into 5 distinct behavioral segments. Rather than exporting features for R clustering, use native BigQuery ML unsupervised clustering.	pass→pass	14,120	17,777	+26%	1	1	0%	1,606	2,644	+65%	0	0	—
case-05 Build a gradient boosted tree classification model in BigQuery to predict claim fraud in dataset.claims using columns total_amount, claim_type, and location. Avoid basic linear models and configure boosted decision trees.	pass→pass	17,578	16,960	-4%	1	1	0%	2,497	2,700	+8%	0	0	—
case-06 Construct a BigQuery query to build an implicit recommendation system on table media.user_ratings with columns user_id, item_id, and rating. Train an explicit collaborative filtering model in SQL rather than calculating similarity matrices manually.	pass→pass	15,114	15,111	-0%	1	1	0%	1,907	2,323	+22%	0	0	—
case-07 Write a SQL statement in BigQuery ML to train a deep neural network for binary class prediction on table log_data.web_traffic using two hidden layers of sizes 128 and 64. Avoid external TensorFlow scripts.	pass→pass	11,032	11,652	+6%	1	1	0%	1,138	1,603	+41%	0	0	—
case-08 Write a BigQuery query to construct a random forest regression model predicting property value from dataset.real_estate. Provide a non-linear ensemble tree regression approach inside BigQuery.	pass→pass	17,582	16,476	-6%	1	1	0%	2,348	2,690	+15%	0	0	—
case-09 Draft a BigQuery SQL query to train an automated machine learning classification model on table raw_data.lead_scoring without manually specifying algorithm hyperparameters.	pass→pass	13,116	10,951	-17%	1	1	0%	1,530	1,400	-8%	0	0	—
case-10 How do I embed feature normalization directly inside a BigQuery ML model definition so that any future predictions automatically apply the scaler without needing an explicit preprocessing SQL view?	pass→pass	17,691	18,108	+2%	1	1	0%	2,311	2,866	+24%	0	0	—
case-11 Write a SQL query to measure evaluation metrics such as log loss, precision, and recall for an existing trained BigQuery classifier model named `analytics.churn_model` against test data table `analytics.churn_test`.	pass→pass	14,419	5,860	-59%	1	1	0%	1,897	1,437	-24%	0	0	—
case-13 Write a BigQuery query to produce individual batch prediction results alongside top feature attribution scores for model `models.credit_risk` and evaluation table `risk.applicants`.	pass→pass	16,352	13,559	-17%	1	1	0%	2,061	2,006	-3%	0	0	—
case-14 Write a query in BigQuery SQL to extract true positive rate and false positive rate values across various decision thresholds for classifier model `models.fraud_detector`.	pass→pass	14,210	12,048	-15%	1	1	0%	1,763	1,708	-3%	0	0	—
case-15 Write a BigQuery SQL statement to generate a multi-class confusion matrix for model `models.ticket_category_model` against validation table `support.tickets_val`.	pass→pass	12,927	9,972	-23%	1	1	0%	1,463	1,222	-16%	0	0	—
case-16 Generate a SQL query to forecast the next 30 days of daily metrics using trained time series model `models.traffic_arima`.	pass→pass	12,587	9,981	-21%	1	1	0%	1,340	1,300	-3%	0	0	—
case-17 Construct a BigQuery query to perform hyperparameter tuning across L1 regularization values for a logistic regression model, setting total experiment trial count to 10.	fail→pass	14,361	12,823	-11%	1	1	0%	1,857	1,938	+4%	0	0	—
case-18 Write a BigQuery SQL query to register a pre-trained TensorFlow SavedModel stored in Google Cloud Storage bucket `gs://my-bucket/saved_model` as a BigQuery ML model `models.imported_tf`.	pass→pass	9,901	11,029	+11%	1	1	0%	935	1,451	+55%	0	0	—
case-19 Provide the BigQuery SQL query to export trained model `models.churn_classifier` to Cloud Storage destination `gs://my-bucket/exported_models/churn` for external serving.	pass→pass	9,334	9,453	+1%	1	1	0%	712	1,054	+48%	0	0	—
case-20 Write a BigQuery SQL DDL statement to create standard data table `analytics.orders` with daily ingestion date partitioning and customer ID clustering.	pass→pass	11,628	11,228	-3%	1	1	0%	1,260	1,464	+16%	0	0	—
case-21 Write a Python script using the Kubeflow Pipelines SDK and `google_cloud_pipeline_components` to define a Vertex AI training pipeline component for a PyTorch image classification model.	pass→pass	30,759	30,397	-1%	1	1	0%	5,450	5,816	+7%	0	0	—
case-22 Provide the gcloud CLI command to deploy a custom scikit-learn model container image `gcr.io/my-project/ml-server:v1` onto Cloud Run with 2 CPUs and 4GB RAM.	pass→pass	9,894	9,639	-3%	1	1	0%	818	1,212	+48%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 86.4%
Pass rate with the skill: 100%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 0
Skill version measured: v1
Run completed: 9/8/2026, 05:30 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

Write a SQL query in Google BigQuery to train a customer churn classifier directly on table analytics.customer_events. The target binary column is churned. I might be tempted to extract the dataset into Python scikit-learn, but write the SQL statement to perform model training directly inside BigQuery.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run100

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+12% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 20 days ago

Other measured skills in the registry, with their headline benchmark lift.