Using JSON - Amazon Kendra

Using JSON

The following examples show how to use JSON for user access control when you create an index.

Warning

The JSON token is a non-validated payload. This should only be used when requests to Amazon Kendra come from a trusted server and never from a browser.

Console

  1. Choose Create index to start creating a new index.

  2. On the Specify index details page, give your index a name and a description.

  3. For IAM role, select a role or select Create a new role to and specify a role name to create a new role. The IAM role will have the prefix "AmazonKendra-".

  4. Leave all of the other fields at their defaults. Choose Next.

  5. In the Configure user access control page, under Access control settings, choose Yes to use tokens for access control.

  6. Under Token configuration, select JSON as the Token type.

  7. Specify a User name to use in the ACL check.

  8. Specify one or more Groups to use in the ACL check.

  9. Choose Next.

  10. In the Provisioning details page, choose Developer edition.

  11. Choose Create to create your index.

  12. Wait for your index to be created. Amazon Kendra provisions the hardware for your index. This operation can take some time.

CLI

To create an index with the AWS CLI using a JSON input file, first create a JSON file with your desired parameters:

{
    "Name": "user-context",
    "Edition": "ENTERPRISE_EDITION",
    "RoleArn": "arn:aws:iam::account-id:role:/my-role",
    "UserTokenConfigurations": [
        {
            "JsonTokenTypeConfiguration": {
                "UserNameAttributeField": "user",
                "GroupAttributeField": "group"
            }
        }
    ],
    "UserContextPolicy": "USER_TOKEN"
}

Next, call create-index using the input file. For example, if the name of your JSON file is create-index-openid.json, you can use the following: 

aws kendra create-index --cli-input-json file://create-index-openid.json

If you are not using Open ID for AWS IAM Identity Center, you can send us the token in JSON format. If you do, you must specify which field in the JSON token contains the user name and which field contains the groups. The group field values must be a JSON string array. For example, if you are using SAML, your token would be similar to the following:

{
     "username" : "user1", 
     "groups": [
        "group1", 
        "group2"
     ]
}

The TokenConfiguration would specify the user name and group field names:

{
    "UserNameAttributeField":"username",
    "GroupAttributeField":"groups"
}
Python
response = kendra.create_index(
    Name='user-context',
    Edition='ENTERPRISE_EDITION',
    RoleArn='arn:aws:iam::account-id:role:/my-role',
    UserTokenConfigurations=[
        {
            "JwtTokenTypeConfiguration": {
                "UserNameAttributeField": "user",
                "GroupAttributeField": "group",
            }
        }
    ],
    UserContextPolicy='USER_TOKEN'
)