Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/altimateai/data-engineering-skills/optimizing-query-textnpx skills add AltimateAI/data-engineering-skills --skill optimizing-query-textgit clone --depth 1 https://github.com/AltimateAI/data-engineering-skillsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00113 | $0.01729 |
| Opus 5 | $0.00056 | $0.00864 |
| Sonnet 5 | $0.00023 | $0.00346 |
| Haiku 4.5 | $0.00011 | $0.00173 |
Grade A, and why
optimizing-query-text scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Optimize Query from SQL Text
OUTPUT FORMAT
Return ONLY the optimized SQL query. No markdown formatting, no explanations, no bullet points - just pure SQL that can be executed directly in Snowflake.
CRITICAL: Semantic Preservation Rules
The optimized query MUST return IDENTICAL results to the original.
Before returning ANY optimization, verify:
- Same columns: Exact same columns in exact same order with exact same aliases
- Same rows: Filter conditions must be semantically equivalent
- Same ordering: Preserve
ORDER BYexactly as written - Same limits: If original has
LIMIT N, keepLIMIT N. If no LIMIT, do NOT add one.
If you cannot guarantee identical results, return the original query unchanged.
Pattern 1: Function on Filter Column
Problem: Functions on columns in WHERE clause prevent partition pruning and index usage.
CAN Fix
| Original | Optimized | Why Safe |
|---|---|---|
WHERE DATE(ts) = '2024-01-01' |
WHERE ts >= '2024-01-01' AND ts < '2024-01-02' |
Equivalent range |
WHERE YEAR(dt) = 2024 |
WHERE dt >= '2024-01-01' AND dt < '2025-01-01' |
Equivalent range |
WHERE MONTH(dt) = 3 AND YEAR(dt) = 2024 |
WHERE dt >= '2024-03-01' AND dt < '2024-04-01' |
Equivalent range |
WHERE DATE(ts) >= '2024-01-01' AND DATE(ts) < '2024-02-01' |
WHERE ts >= '2024-01-01' AND ts < '2024-02-01' |
Same boundaries |
WHERE YEAR(dt) BETWEEN 1995 AND 1996 |
WHERE dt >= '1995-01-01' AND dt < '1997-01-01' |
Equivalent range |
CANNOT Fix
| Pattern | Why Not |
|---|---|
WHERE YEAR(dt) IN (SELECT year FROM ...) |
Dynamic values, cannot precompute range |
WHERE DATE(ts) = DATE(other_col) |
Comparing two columns, both need function |
WHERE EXTRACT(DOW FROM dt) = 1 |
Day-of-week has no contiguous range |
WHERE DATE_TRUNC('month', dt) = '2024-01-01' in GROUP BY |
Needed for grouping logic |
SELECT YEAR(dt) AS yr ... GROUP BY YEAR(dt) |
Function in SELECT/GROUP BY is fine, only filter matters |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 167 lines · 113 tokens per session scan A 5007588fe49f
optimizing-query-text is a skill published in the GitHub repository AltimateAI/data-engineering-skills (122 stars, last pushed 1mo ago), licensed MIT. It adds 113 tokens to every session and 1,729 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
explain-sql
Explain complex SQL queries in plain English with Mermaid data flow diagrams, performance annotations, and anti-pattern detection. Use when the user pastes a SQL query and asks "what does this do?", "explain this query", or needs to understand inherited SQL, CTEs, window functions, recursive queries, or dbt model…
sql-optimizer
Analyze and optimize slow SQL queries. Use when the user says a query is slow, asks to optimize or speed up SQL, wants to find anti-patterns, needs index recommendations, or asks for a query rewrite. Also use when EXPLAIN output shows full table scans or poor join strategies.
aegis-dq
Agentic data quality validation across warehouses (DuckDB, BigQuery, Athena, Databricks, Postgres) with LLM diagnosis, root cause analysis, and audit trail.
data-parity
Validate that two tables or query results are identical — or diagnose exactly how they differ. Discover schema, identify keys, profile cheaply, then diff. Use for migration validation, ETL regression, and query refactor verification.
dbt-develop
REQUIRED before writing or modifying ANY dbt model. Invoke this skill FIRST whenever a task says "create", "build", "add", "modify", "update", "fix", or "refactor" a dbt model, staging file, mart, incremental, or snapshot. Skipping this skill is the leading cause of silent-correctness bugs — models that compile and…
dbt-troubleshoot
Debug dbt errors — compilation failures, runtime database errors, test failures, wrong data, and performance issues. Use when something is broken, producing wrong results, or failing to build. Powered by altimate-dbt.