Instead of parsing JSON output from AWS ECR commands, you can format the results as a readable table using the --output table flag.

Default JSON output:

aws ecr describe-images --repository-name my-repo --region us-east-1
# Returns verbose JSON

Table output:

aws ecr describe-images \
  --repository-name my-repo \
  --region us-east-1 \
  --query 'sort_by(imageDetails,&imagePushedAt)[-1]' \
  --output table

What you get:

Instead of JSON blobs, you’ll see a cleanly formatted table with:

• Column headers
• Aligned values
• Much better readability
• Easy scanning of image details

Other useful –output options:

• table - Human-readable table format
• text - Tab-delimited text
• json - Default JSON format (verbose)
• yaml - YAML format

Combining with –query:

The --query parameter lets you filter and sort results using JMESPath:

# Get the most recently pushed image
--query 'sort_by(imageDetails,&imagePushedAt)[-1]'

# Get only specific fields
--query 'imageDetails[*].[imageTags[0],imagePushedAt]'

This makes it much easier to quickly check image details, tags, and push dates without writing JSON parsing scripts.

Original source