# review

Docs for TrueQuery 0.1.6. Run every check, then the receipt and a critique of what checks can't see.

**For everyone, before you merge.** In Claude Code this is a `/truequery` slash command. In Muse Code, describe the task in your own words, or load the skill with `/skill truequery`. In Codex the skill isn't here yet, only the checks.

```
/truequery review models/
```

## What it does

1. **Runs every check** and quotes the receipt. A pass with files it couldn't check isn't a pass, and it says which ones.
2. **Deals with each finding:** it fixes it, or adds the narrowest ignore with a reason.
3. **Reviews what checks can't see.** It reads the whole path, compares your team's definitions with the code, and looks at grain, units, time zones, filters and NULLs.
4. **Confirms in the data** when it can reach it read-only, and labels each finding "confirmed in data" or "by reading".
5. **Reports** blockers first, then should-fix, then minor, and ends with a merge recommendation.

## The checker on its own

Without an agent, after `pip install truequery`, `truequery review` gives you the findings and the receipt, not the critique. `⋯` marks the findings, cut here:

```output
$ truequery review samples/bad_orders.sql
⋯
TrueQuery receipt
Checked      1 SQL file · contracts for 0 of 3 tables
Found        5 errors · 2 warnings
Not checked  your data: TrueQuery reads files and never connects to your database
No contract  customers, orders, staging_orders
```
