Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill should be used when the user asks to "optimize Snowflake queries", "analyze Snowflake SQL performance", "size Snowflake warehouses", "review Snowflake data models", or "troubleshoot Snowflake cost issues".
.claude/skills/borghei-snowflake-development/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 274% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 184% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -1% | 0% |
> Category: Engineering > Domain: Data Warehouse
The Snowflake Development skill provides tools for analyzing and optimizing Snowflake SQL queries, recommending warehouse sizing, and enforcing Snowflake-specific best practices. Helps data engineers reduce costs and improve query performance.
Before analyzing or sizing, confirm these inputs. If any is unknown or vague, ASK — do not assume:
--action; selects the workflow)--file; the subject of the analysis)--workload/--data-volume; drives the warehouse recommendation)Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
bash# Analyze a Snowflake SQL file for optimization opportunities python scripts/snowflake_query_helper.py --file queries.sql --action analyze # Get warehouse sizing recommendations python scripts/snowflake_query_helper.py --action warehouse-sizing --workload "etl" --data-volume "500GB" # Optimize a specific query python scripts/snowflake_query_helper.py --file slow_query.sql --action optimize
| Tool | Purpose | Key Flags | |------|---------|-----------| | snowflake_query_helper.py | Analyze, optimize Snowflake SQL and recommend warehouse sizes | --file, --action, --workload, --data-volume |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 5,359 | 14,324 | +167% | 1 | 1 | 0% | 791 | 2,959 | +274% | 0 | 0 | — |
case-02 | fail→pass | 5,771 | 9,680 | +68% | 1 | 1 | 0% | 764 | 2,171 | +184% | 0 | 0 | — |
case-03 | pass→pass | 13,395 | 9,714 | -27% | 1 | 1 | 0% | 2,354 | 2,176 | -8% | 0 | 0 | — |
case-04 | pass→pass | 11,217 | 8,863 | -21% | 1 | 1 | 0% | 1,931 | 1,965 | +2% | 0 | 0 | — |
case-05 | fail→fail | 11,683 | 14,518 | +24% | 1 | 1 | 0% | 1,980 | 3,420 | +73% | 0 | 0 | — |
case-06 | fail→pass | 9,005 | 2,966 | -67% | 1 | 1 | 0% | 1,315 | 1,041 | -21% | 0 | 0 | — |
case-07 | fail→pass | 12,983 | 3,482 | -73% | 1 | 1 | 0% | 1,843 | 1,128 | -39% | 0 | 0 | — |
case-08 | pass→pass | 5,288 | 1,965 | -63% | 1 | 1 | 0% | 709 | 856 | +21% | 0 | 0 | — |
case-09 | pass→pass | 13,460 | 3,274 | -76% | 1 | 1 | 0% | 1,970 | 1,013 | -49% | 0 | 0 | — |
case-10 | pass→pass | 10,394 | 9,578 | -8% | 1 | 1 | 0% | 1,528 | 1,905 | +25% | 0 | 0 | — |
case-11 | fail→pass | 13,761 | 9,898 | -28% | 1 | 1 | 0% | 2,021 | 2,000 | -1% | 0 | 0 | — |
case-12 | pass→pass | 17,255 | 14,028 | -19% | 1 | 1 | 0% | 2,596 | 2,663 | +3% | 0 | 0 | — |
case-13 | pass→pass | 11,331 | 6,063 | -46% | 1 | 1 | 0% | 1,616 | 1,438 | -11% | 0 | 0 | — |
case-14 | pass→pass | 11,401 | 11,453 | +0% | 1 | 1 | 0% | 1,668 | 1,867 | +12% | 0 | 0 | — |
case-15 | pass→pass | 4,162 | 1,985 | -52% | 1 | 1 | 0% | 533 | 863 | +62% | 0 | 0 | — |
case-16 | fail→pass | 7,509 | 1,779 | -76% | 1 | 1 | 0% | 1,024 | 791 | -23% | 0 | 0 | — |
case-17 | fail→pass | 7,900 | 2,004 | -75% | 1 | 1 | 0% | 1,218 | 856 | -30% | 0 | 0 | — |
case-18 | pass→pass | 12,000 | 8,331 | -31% | 1 | 1 | 0% | 1,929 | 1,834 | -5% | 0 | 0 | — |
case-19 | fail→pass | 6,476 | 2,516 | -61% | 1 | 1 | 0% | 1,009 | 970 | -4% | 0 | 0 | — |
case-20 | fail→pass | 12,660 | 3,133 | -75% | 1 | 1 | 0% | 1,825 | 1,073 | -41% | 0 | 0 | — |
case-21 | pass→pass | 7,005 | 3,063 | -56% | 1 | 1 | 0% | 1,030 | 1,050 | +2% | 0 | 0 | — |
case-22 | fail→pass | 10,036 | 2,680 | -73% | 1 | 1 | 0% | 1,457 | 927 | -36% | 0 | 0 | — |
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 +45 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.
Other measured skills in the registry, with their headline benchmark lift.