Skip to main content
PawSQL SQL Review applies database-aware static analysis and policy checks before SQL reaches a target environment. Each review produces structured findings with severity, rule context, affected objects, and recommended action—giving developers, DBAs, and release teams a shared basis for remediation and approval.

​
Review lifecycle

A ticket moves through the following states from creation to execution: To complete your first review, follow these operating steps (the first three prepare the inputs to ticket creation):
1

Define the review target

Identify the database engine, version, project, and environment where the SQL will run.
2

Choose a policy

Select a policy built for the target database and verify its rules, severities, and thresholds.
3

Prepare SQL and context

Paste statements or upload a script, and attach a workspace when object metadata is needed.
4

Create the review ticket

Fill in the title and priority, submit, and wait for the automated review to finish.
5

Resolve and disposition findings

Address the highest-risk items first, run the review again, and document any accepted exceptions.
6

Submit for review and approval

Submit the ticket for human review, have an approver decide, and track execution through the execution records after approval.

​
Key concepts

This guide uses review policy throughout. The interface’s navigation and form display review template, and the template page’s create button reads create rule template—all the same reusable collection of rules, severities, thresholds, and exceptions.

​
In this section

​
Run a review

Create a review ticket

Read review results

Resolve findings

Manual review and approval

History, reports, and export

​
Configuration and governance

Review policy

Workspace and review context

Control high-risk SQL

​
Before you begin

  • Confirm the target database engine and version.
  • Use a review policy designed for that database.
  • Verify script encoding, dialect, and statement delimiters.
  • Prepare a workspace when the review depends on object metadata.
  • Remove or mask sensitive values in SQL, DDL, comments, and logs.
  • Confirm that your account can create and view review tickets.
Automated review improves coverage and consistency. It does not replace business validation, testing, backup planning, change approval, or rollback preparation.