> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pawsql.com/llms.txt
> Use this file to discover all available pages before exploring further.
> ## Agent Instructions
> PawSQL 是一个产品：Cloud 是公网部署形态，Engine / Optimizer / Auditor / Advisor / Patroller 是同一产品的组件与交付形态，不是彼此独立的产品。 / PawSQL is a single product: Cloud is the public deployment form, while Engine / Optimizer / Auditor / Advisor / Patroller are components and delivery forms of the same product, not separate products.
> 术语以站内术语表为准：SQL 审核对应英文 SQL Review，查询重写对应 Query Rewrite，索引推荐对应 Index Recommendation；英文内容统一用 Review，不用 Audit。 / Use the site glossary for terminology: 审核 is SQL Review, 重写 is Query Rewrite, 索引推荐 is Index Recommendation; English content uses Review, never Audit.
> 引用能力范围或版本支持时以对应页面为准；标注 unknown、或 status 非 published 的内容表示尚未经产品核实，不应作为事实引用。 / Cite capability scope and version support from the corresponding page; content marked unknown, or with a status other than published, is not yet product-verified and must not be cited as fact.
# Developer SQL Copilot
> Analyze, optimize, and validate SQL while you code. PawSQL helps developers detect performance risks and fix them before SQL reaches production.
## Audience
Backend developers (Java, Go, and similar) who write SQL every day but don't need to become database optimization experts.
## Problem
A query that returns in milliseconds in development (tens of thousands of rows) can become a bottleneck in production (tens to hundreds of millions of rows).
Consider an order query:
```sql theme={null}
SELECT order_id, customer_id, order_status, create_time, total_amount
FROM orders
WHERE DATE(create_time) = '2026-09-04'
AND customer_id = 10086
ORDER BY create_time DESC;
```
`DATE(create_time)` wraps the filter column in a function, preventing the optimizer from efficiently using an index on `create_time` and widening the scan.
Developers typically don't check for these issues while writing SQL — execution plans, index structure, optimizer behavior, and predicate selectivity aren't part of their daily work.
## Goal
Find and fix SQL performance risks before the code is committed, so problems stop at development instead of surfacing in production months later.
## Workflow
Developer SQL Copilot brings database performance analysis directly into the development workflow:
```mermaid theme={null}
flowchart TD
A[Write SQL] --> B[Analyze SQL]
B --> C[Detect Performance Risk]
C --> D[Explain Why]
D --> E[Generate Equivalent Rewrite]
E --> F[Recommend Index]
F --> G[Validate Execution Plan]
G --> H[Apply Fix Before Commit]
```
The whole flow happens inside the existing development process — developers don't need to switch tools.
## Dependencies
Built from the following capabilities (see each capability page for rewrite rules and index-design logic):
## Success Criteria
| Metric | Objective |
| ---------------------------------------------- | --------- |
| SQL issues found during development | Increase |
| SQL issues discovered after deployment | Decrease |
| Average SQL optimization turnaround time | Decrease |
| Developer-to-DBA consultation requests | Decrease |
| Production slow queries caused by new releases | Decrease |
***
## Beyond Generating SQL
Many AI coding assistants focus on "how do I write this SQL?". But production-grade SQL also needs to answer:
> Will this use the right index? Will it still perform on a hundred million rows? Is there a better rewrite? What index should I add? Is the execution plan reasonable? Is it actually faster after optimization?
Developer SQL Copilot's value is here — it doesn't just generate SQL; it analyzes, rewrites, and validates against database semantics and the optimizer.
## Developer SQL Copilot + SQL Quality Gate
Developer SQL Copilot solves the problem at coding time; SQL Quality Gate solves it at commit and release time. Together they form a shift-left performance strategy:
```mermaid theme={null}
flowchart TD
A[Developer] --> B[Write SQL]
B --> C[Developer SQL Copilot]
C --> D["Find and Fix"]
D --> E[Commit]
E --> F[Pull Request]
F --> G[SQL Quality Gate]
G --> H["Verify and Enforce"]
H --> I[Production]
```
Run your first quality check, rewrite, and index recommendation.
Block risky SQL in your CI/CD pipeline.
See how equivalent rewrites are generated.
Talk to the PawSQL team.