---
title: Output and errors
description: Read kapten results as tables or JSON, filter them for scripts, and handle errors and exit codes.
---

## Tables and JSON

Commands that list resources print a table of their main fields. Other commands print the resource as JSON. `-o` chooses the format:

```bash
kapten projects list                 # a table
kapten projects list -o json         # the API's response, pagination included
kapten projects get <project-id> -o yaml
```

`--rsh-columns` chooses a table's columns, and `--rsh-sort-by` the column its rows are sorted by:

```bash
kapten environments list <project-id> --rsh-columns name,type,id --rsh-sort-by name
```

## Filter results

`-f` keeps only part of the response, with a [jq](https://jqlang.org) expression. The response is under `.body`:

```bash
kapten projects list -f '.body.records[].id'
kapten projects get <project-id> -f '.body.name'
```

Lists return one page of results at a time: `--page-limit` sets the page size and `--page-number` the page.

## Errors

When the API refuses a request, the CLI prints the error on one line to stderr and exits with status `1`. stdout only ever holds results, so a script can pipe it safely. With `-o json`, the API's error document is printed to stdout too.

When your session has ended, the error says to run `kapten login` again.

`-v` prints the HTTP requests and responses, without tokens, to help find what went wrong:

```bash
kapten projects list -v
```
