How to save AWS CLI output to a file

Saving AWS CLI output to a file keeps command results available after terminal scrollback is gone and makes it easier to attach evidence to a ticket, compare repeated checks, or hand structured data to another tool without copying from the screen.

The AWS CLI prints normal command results to stdout, and the shell redirection operators >, >>, and pipes decide whether that stdout stream is written to disk, appended to a log, or duplicated with tee. The file contents still depend on CLI-side flags such as --output and --query, so the saved result can stay as json, yaml, or flat text depending on what the next step needs.

Pager, debug, and credential state change what actually reaches the file. --no-cli-pager prevents long output from being sent through the client-side pager, --debug writes diagnostics to stderr instead of stdout, and --output off suppresses stdout entirely. When a profile is not ready for live API calls, --generate-cli-skeleton output is a safe way to rehearse the save workflow locally with AWS-shaped sample output before switching to a real request.

Steps to save AWS CLI output to a file:

Create a directory for saved command results before redirecting output into multiple files.
```
$ mkdir -p ~/aws-output
```
A dedicated directory keeps normal results, appended logs, and debug captures together instead of scattering them across the shell history and home directory.

Save a safe local command result to a file and confirm that redirection is writing stdout where expected.

$ aws configure list --no-cli-pager > ~/aws-output/aws-config.txt
$ sed -n '1,5p' ~/aws-output/aws-config.txt
NAME       : VALUE                    : TYPE             : LOCATION
profile    : <not set>                : None             : None
access_key : ****************MPLE     : shared-credentials-file : 
secret_key : ****************EKEY     : shared-credentials-file : 
region     : us-west-2                : config-file      : ~/.aws/config

--no-cli-pager is a safe default in saved-output examples because it guarantees that the CLI writes directly to stdout instead of opening the client-side pager.

Save structured AWS-shaped output locally when the save workflow needs to be tested without spending an API call or depending on valid credentials.

$ aws ec2 describe-regions --generate-cli-skeleton output \
  --output json --no-cli-pager > ~/aws-output/regions.json
$ cat ~/aws-output/regions.json
{
    "Regions": [
        {
            "OptInStatus": "OptInStatus",
            "Geography": [
                {
                    "Name": "Name"
                }
            ],
            "RegionName": "RegionName",
            "Endpoint": "Endpoint"
        }
    ]
}

--generate-cli-skeleton output returns a sample response locally, so the saved file shows the response shape without contacting AWS. The placeholder values are illustrative only.

Change the saved file format or reduce the saved fields before writing the file when the next tool expects YAML or a flat text list.
```
$ aws ec2 describe-regions --generate-cli-skeleton output \
  --output yaml --no-cli-pager > ~/aws-output/regions.yaml
$ aws ec2 describe-regions --generate-cli-skeleton output \
  --query 'Regions[].RegionName' --output text --no-cli-pager > ~/aws-output/region-names.txt
$ cat ~/aws-output/region-names.txt
RegionName
```
Prefer json or yaml while shaping a live --query expression. AWS documents that --output text paginates first and then runs the query once per page, which can produce repeated or surprising matches on larger live responses.

Related: How to use JMESPath queries in AWS CLI
Related: How to set the default output format in AWS CLI
Apply the same redirection pattern to a real AWS API call once valid credentials are loaded for the target profile.
```
$ aws sts get-caller-identity --output json --no-cli-pager > ~/aws-output/caller-identity.json
```
Live service commands save only stdout with >. If the request fails because of an expired or invalid token, the error stays on stderr and the redirected file does not contain the JSON response.

Related: How to check the current caller identity in AWS CLI
Related: How to set the default output format in AWS CLI
Use tee when the same result should stay visible in the terminal and be written to a file at the same time.
```
$ aws ec2 describe-regions --generate-cli-skeleton output \
  --query 'Regions[].RegionName' --output text --no-cli-pager | tee ~/aws-output/region-names.txt
RegionName
```
tee is useful during manual checks because the saved file and the on-screen result stay in sync without running the command twice.

Append repeated output with >> when the file should act as a running log instead of being replaced on each command.

$ aws ec2 describe-regions --generate-cli-skeleton output \
  --query 'Regions[].RegionName' --output text --no-cli-pager >> ~/aws-output/region-history.log
$ tail -n 3 ~/aws-output/region-history.log
RegionName
RegionName
RegionName

Use >> only when repeated lines are expected. Use > when the file should contain only the most recent result.

Save normal output and debug logs into separate files when the command result must be archived without mixing in diagnostic noise.
```
$ aws ec2 describe-regions --generate-cli-skeleton output \
  --debug --no-cli-pager > ~/aws-output/regions.json 2> ~/aws-output/regions-debug.log
$ wc -c < ~/aws-output/regions.json
277
$ wc -c < ~/aws-output/regions-debug.log
16581
```
AWS documents --debug as stderr diagnostic output and recommends redirecting it to a file when the log needs to be searched later.
Avoid --output off when the goal is to save the response body, because that output mode suppresses stdout completely.
```
$ aws ec2 describe-regions --generate-cli-skeleton output \
  --output off --no-cli-pager > ~/aws-output/regions-empty.txt
$ wc -c < ~/aws-output/regions-empty.txt
0
```
--output off is useful only when the exit status matters more than the response body. With > redirection, it creates an empty file instead of saving the command result.