Ready to supercharge your AI agents?

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

Sql Transform Helper · dicklesworthstone | DecimalAI

dicklesworthstone/sql-transform-helper

NewSkillScoreMaintained

Assist with sql transform helper operations. Auto-activating skill for Data Pipelines. Triggers on: sql transform helper, sql transform helper Part of the Data Pipelines skill category. Use when working with sql transform helper functionality. Trigger with phrases like "sql transform helper", "sql helper", "sql".

by dicklesworthstone

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

Automatic

Data Engineer pack

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

pip install decimalai

2 · Get the skill — writes .claude/skills/dicklesworthstone-sql-transform-helper/SKILL.md

decimalai skills pull dicklesworthstone-sql-transform-helper --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/dicklesworthstone-sql-transform-helper/SKILL.md https://app.decimal.ai/s/dicklesworthstone-sql-transform-helper/SKILL.md

Raw file: /s/dicklesworthstone-sql-transform-helper/SKILL.md ↗ · machine-readable summary at /s/dicklesworthstone-sql-transform-helper.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 23d ago

No measurable change+24% tokens0% turns91% of whole cases pass with skill (20/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-19	✗→✓	▲ Improved	12%	0%
case-17	✓→✗	▼ Worse	27%	0%
case-01	✓→✓	＝ Same ✓	64%	0%
case-02	✓→✓	＝ Same ✓	63%	0%
case-03	✓→✓	＝ Same ✓	59%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

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/11-data-pipelines/sql-transform-helper/SKILL.md

View on GitHub ↗

License MITImported 8/10/2026Version v1

Skill contents

SKILL.md

Sql Transform Helper

Overview

This skill provides automated assistance for sql transform helper tasks within the Data Pipelines domain.

When to Use

This skill activates automatically when you:

Mention "sql transform helper" in your request
Ask about sql transform helper patterns or best practices
Need help with data pipeline skills covering etl, data transformation, workflow orchestration, and streaming data processing.

Instructions

Provides step-by-step guidance for sql transform helper
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 sql transform helper" Result: Provides step-by-step guidance and generates appropriate configurations

Prerequisites

Relevant development environment configured
Access to necessary tools and services
Basic understanding of data pipelines 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 Data Pipelines skill category. Tags: etl, airflow, spark, streaming, data-engineering

Illustrative Want to see what dicklesworthstone/sql-transform-helper 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

8/30/2026, 17:29 UTC · 22 cases

Lift

91% vs 91%

Agent turns

1.0 vs 1.0

Tokens

+24%

2,664.23 vs 2,143

fail → pass · 1pass → pass · 19fail → fail · 1pass → fail · 1

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-01 Write a data transformation query in SQL to deduplicate events by user_id, keeping only the single latest record according to event_timestamp. A team member suggested using GROUP BY user_id, event_timestamp with MAX(event_timestamp), but that loses non-aggregated columns. Provide the window function approach.	pass→pass	7,741	14,543	+88%	1	1	0%	1,288	2,115	+64%	0	0	—
case-02 We need to update a dimensional customer table maintaining history via SCD Type 2 from an incoming staging table stg_customers. A developer proposed running UPDATE dim_customers SET status = 'INACTIVE' to mark old records. Write a SQL MERGE query pattern that properly updates valid_to on existing records and inserts new versions.	pass→pass	20,963	24,267	+16%	1	1	0%	2,993	4,888	+63%	0	0	—
case-03 In a PySpark Structured Streaming pipeline processing sensor records, late-arriving events up to 10 minutes past event_timestamp must be included before computing 5-minute tumbling aggregations. Relying only on processing-time triggers drops late data. Provide the PySpark Structured Streaming transformation code.	pass→pass	17,616	18,831	+7%	1	1	0%	1,877	2,983	+59%	0	0	—
case-04 Write a dbt model query for processing transaction logs incrementally based on loaded_at. Omission of incremental materialization conditions causes full table scans on every run. Provide the dbt SQL template with Jinja incremental filtering.	pass→pass	16,479	13,767	-16%	1	1	0%	1,705	1,915	+12%	0	0	—
case-05 We have a monthly sales table with columns region, month_name, and revenue. Convert this into a quarterly reporting table with Q1_rev, Q2_rev, Q3_rev, and Q4_rev columns per region. Avoid writing four separate LEFT JOIN subqueries on the same table.	pass→pass	8,603	15,500	+80%	1	1	0%	1,690	2,462	+46%	0	0	—
case-06 In a Spark SQL transformation pipeline, we join a 10TB transactions table with a 5MB store lookup table. The execution plan defaults to a full shuffle hash join across all nodes. Write the Spark SQL query optimized to eliminate network shuffle overhead.	pass→pass	12,577	24,132	+92%	1	1	0%	1,286	1,710	+33%	0	0	—
case-07 Track user continuous sessions in SQL where a new session ID starts after 30 minutes of inactivity between events. A naive implementation uses nested scalar subqueries for each row. Provide the window function approach using offset checking and running totals.	pass→pass	12,036	17,257	+43%	1	1	0%	2,472	2,750	+11%	0	0	—
case-08 We are joining staging_accounts and main_accounts on composite keys account_id and sub_account_id, where sub_account_id can contain NULL values that must match NULLs in the target table. Standard '=' join conditions drop rows with NULL keys. How should the join condition be written in standard SQL?	pass→pass	15,513	17,685	+14%	1	1	0%	1,850	2,264	+22%	0	0	—
case-09 Transform a wide survey table with columns customer_id, score_q1, score_q2, score_q3 into a tall schema with customer_id, question_id, and score columns. Avoid writing multiple UNION ALL queries that scan the base table repeatedly.	pass→pass	16,491	19,245	+17%	1	1	0%	2,338	2,598	+11%	0	0	—
case-10 In Spark SQL, incoming JSON user order records contain a nested array field named items_purchased. Expand each element of this array into individual rows while keeping top-level order_id fields intact.	pass→pass	15,667	8,586	-45%	1	1	0%	2,105	2,119	+1%	0	0	—
case-11 We need a report showing total daily active users for every date in 2024, including days with zero activity. Joining directly against the user_activity log table skips dates without events. Provide the SQL pattern using continuous date generation.	pass→pass	13,641	13,726	+1%	1	1	0%	2,105	2,426	+15%	0	0	—
case-12 Calculate the median support ticket resolution time (in seconds) per support team queue in SQL. Using AVG(resolution_seconds) produces incorrect results due to outlier skew. Write the exact median calculation.	pass→pass	10,889	11,538	+6%	1	1	0%	2,199	2,576	+17%	0	0	—
case-13 Configure an Apache Airflow DAG task to run a PySpark data transformation job on a cluster, passing the dynamic execution date {{ ds }} into application arguments. What standard Airflow operator handles PySpark application submission?	pass→pass	12,845	19,033	+48%	1	1	0%	2,026	2,806	+38%	0	0	—
case-14 Compute a daily cumulative running sum of sales revenue per store for fiscal year 2024. A self-join query matching date >= date scales with O(N^2) complexity. Write the O(N) SQL window function query.	pass→pass	10,867	14,712	+35%	1	1	0%	1,534	2,226	+45%	0	0	—
case-15 In an ETL pipeline processing Kafka CDC messages from Debezium with operation flags ('c', 'u', 'd') and sequence number pos, reconstruct the current state of each record by applying updates and purging deleted rows.	pass→pass	17,730	21,322	+20%	1	1	0%	2,927	3,477	+19%	0	0	—
case-16 A Spark SQL join between order_facts and tenant_dim suffers severe data skew on key tenant_id = 0, causing executor memory OOM errors. Explain and write the SQL key salting transform pattern to distribute skewed keys across nodes.	pass→pass	21,514	23,806	+11%	1	1	0%	3,134	3,852	+23%	0	0	—
case-17 Generate a unique composite primary key in a dbt model from tenant_id, user_id, and subscription_code while handling NULLs safely. Manual string concatenation with \|\| produces invalid keys when columns contain NULL.	pass→fail	16,689	47,098	+182%	1	1	0%	1,962	2,499	+27%	0	0	—
case-18 When running a daily ETL pipeline backfill for date partition ds = '2024-01-15' in Spark SQL, standard INSERT INTO appends duplicate rows. How do you configure and write the statement to replace only the target partition?	pass→pass	17,366	16,329	-6%	1	1	0%	2,259	2,474	+10%	0	0	—
case-19 Write a SQL MERGE query to sync target_users with stg_users. The logic must update modified details, insert non-existent rows, and delete target records when stg_users.is_deleted = TRUE.	fail→pass	14,326	8,694	-39%	1	1	0%	1,784	2,000	+12%	0	0	—
case-20 We are designing an OLTP MySQL database schema for high-throughput transactional order lookups and range scans on created_at. Should we use Hash indexes or B-Tree indexes on InnoDB storage engine tables?	pass→pass	19,099	17,903	-6%	1	1	0%	2,269	2,874	+27%	0	0	—
case-21 Our Apache Kafka cluster nodes experience high garbage collection pause times under heavy streaming load. How should we set initial (-Xms) and maximum (-Xmx) JVM heap sizes in kafka-env.sh for optimal broker stability?	pass→pass	23,502	18,501	-21%	1	1	0%	2,654	2,629	-1%	0	0	—
case-22 Our Airflow cluster uses CeleryExecutor on worker nodes with 8 CPU cores and 32GB RAM. How should we tune worker_concurrency in airflow.cfg to prevent worker OOM crashes during heavy parallel tasks?	fail→fail	20,546	19,904	-3%	1	1	0%	2,689	2,970	+10%	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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 90.9%
Pass rate with the skill: 90.9%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 8/30/2026, 17:29 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-19

Prompt

Write a SQL MERGE query to sync target_users with stg_users. The logic must update modified details, insert non-existent rows, and delete target records when stg_users.is_deleted = TRUE.

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 signalsno signals yet

BenchmarkRun a verified benchmark with positive lift

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+26% tokens

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

No signals yet — run a benchmark or collect evaluated traces to earn a score.

Verified on gemini-3.6-flash · re-tested 22 days ago

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