Skip to content
|
Plugin Hub
github-edit-pageEdit this page
report-issueReport an issue
header icon

Basic config examples for AI Response Transformer

By Kong Inc.

Basic configuration examples

The following examples provide some typical configurations for enabling the ai-response-transformer plugin on a service.

Make the following request:

curl -X POST http://localhost:8001/services/{serviceName|Id}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --data '
    {
  "name": "ai-response-transformer",
  "config": {
    "prompt": "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.",
    "transformation_extract_pattern": "\\\\{((.|\\n)*)\\\\}",
    "parse_llm_response_json_instructions": false,
    "llm": {
      "route_type": "llm/v1/chat",
      "auth": {
        "header_name": "api-key",
        "header_value": "<AZURE_OPENAI_TOKEN>"
      },
      "logging": {
        "log_statistics": true,
        "log_payloads": false
      },
      "model": {
        "provider": "azure",
        "name": "gpt-35-turbo",
        "options": {
          "max_tokens": 1024,
          "temperature": 1.0,
          "azure_instance": "azure-openai-instance-name",
          "azure_deployment_id": "gpt-3-5-deployment"
        }
      }
    }
  }
}
    '

Replace SERVICE_NAME|ID with the id or name of the service that this plugin configuration will target.

Make the following request, substituting your own access token, region, control plane ID, and service ID:

curl -X POST \
https://{us|eu}.api.konghq.com/v2/control-planes/{controlPlaneId}/core-entities/services/{serviceId}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --header "Authorization: Bearer TOKEN" \
    --data '{"name":"ai-response-transformer","config":{"prompt":"For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.","transformation_extract_pattern":"\\\\{((.|\\n)*)\\\\}","parse_llm_response_json_instructions":false,"llm":{"route_type":"llm/v1/chat","auth":{"header_name":"api-key","header_value":"<AZURE_OPENAI_TOKEN>"},"logging":{"log_statistics":true,"log_payloads":false},"model":{"provider":"azure","name":"gpt-35-turbo","options":{"max_tokens":1024,"temperature":1.0,"azure_instance":"azure-openai-instance-name","azure_deployment_id":"gpt-3-5-deployment"}}}}}'

See the Konnect API reference to learn about region-specific URLs and personal access tokens.

First, create a KongPlugin resource:

echo "
apiVersion: configuration.konghq.com/v1
kind: KongPlugin
metadata:
  name: ai-response-transformer-example
plugin: ai-response-transformer
config:
  prompt: For any city name, put the country that it's in, in brackets next to it. Reply
    with only the JSON result.
  transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
  parse_llm_response_json_instructions: false
  llm:
    route_type: llm/v1/chat
    auth:
      header_name: api-key
      header_value: "<AZURE_OPENAI_TOKEN>"
    logging:
      log_statistics: true
      log_payloads: false
    model:
      provider: azure
      name: gpt-35-turbo
      options:
        max_tokens: 1024
        temperature: 1.0
        azure_instance: azure-openai-instance-name
        azure_deployment_id: gpt-3-5-deployment
" | kubectl apply -f -

Next, apply the KongPlugin resource to an ingress by annotating the service as follows:

kubectl annotate service SERVICE_NAME konghq.com/plugins=ai-response-transformer-example

Replace SERVICE_NAME with the name of the service that this plugin configuration will target. You can see your available ingresses by running kubectl get service.

Note: The KongPlugin resource only needs to be defined once and can be applied to any service, consumer, or route in the namespace. If you want the plugin to be available cluster-wide, create the resource as a KongClusterPlugin instead of KongPlugin.

Add this section to your declarative configuration file:

plugins:
- name: ai-response-transformer
  service: SERVICE_NAME|ID
  config:
    prompt: For any city name, put the country that it's in, in brackets next to it. Reply
      with only the JSON result.
    transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
    parse_llm_response_json_instructions: false
    llm:
      route_type: llm/v1/chat
      auth:
        header_name: api-key
        header_value: "<AZURE_OPENAI_TOKEN>"
      logging:
        log_statistics: true
        log_payloads: false
      model:
        provider: azure
        name: gpt-35-turbo
        options:
          max_tokens: 1024
          temperature: 1.0
          azure_instance: azure-openai-instance-name
          azure_deployment_id: gpt-3-5-deployment

Replace SERVICE_NAME|ID with the id or name of the service that this plugin configuration will target.

Prerequisite: Configure your Personal Access Token 
terraform {
  required_providers {
    konnect = {
      source  = "kong/konnect"
    }
  }
}

provider "konnect" {
  personal_access_token = "kpat_YOUR_TOKEN"
  server_url            = "https://us.api.konghq.com/"
}

Add the following to your Terraform configuration to create a Konnect Gateway Plugin:

resource "konnect_gateway_plugin_ai_response_transformer" "my_ai_response_transformer" {
  enabled = true

  config = {
    prompt = "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result."
    transformation_extract_pattern = "\\{((.|\n)*)\\}"
    parse_llm_response_json_instructions = false
    llm = {
      route_type = "llm/v1/chat"
      auth = {
        header_name = "api-key"
        header_value = "<AZURE_OPENAI_TOKEN>"
      }
      logging = {
        log_statistics = true
        log_payloads = false
      }
      model = {
        provider = "azure"
        name = "gpt-35-turbo"
        options = {
          max_tokens = 1024
          temperature = 1.0
          azure_instance = "azure-openai-instance-name"
          azure_deployment_id = "gpt-3-5-deployment"
        }
      }
    }
  }

  control_plane_id = konnect_gateway_control_plane.my_konnect_cp.id
  service = {
    id = konnect_gateway_service.my_service.id
  }
}

The following examples provide some typical configurations for enabling the ai-response-transformer plugin on a route.

Make the following request:

curl -X POST http://localhost:8001/routes/{routeName|Id}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --data '
    {
  "name": "ai-response-transformer",
  "config": {
    "prompt": "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.",
    "transformation_extract_pattern": "\\\\{((.|\\n)*)\\\\}",
    "parse_llm_response_json_instructions": false,
    "llm": {
      "route_type": "llm/v1/chat",
      "auth": {
        "header_name": "api-key",
        "header_value": "<AZURE_OPENAI_TOKEN>"
      },
      "logging": {
        "log_statistics": true,
        "log_payloads": false
      },
      "model": {
        "provider": "azure",
        "name": "gpt-35-turbo",
        "options": {
          "max_tokens": 1024,
          "temperature": 1.0,
          "azure_instance": "azure-openai-instance-name",
          "azure_deployment_id": "gpt-3-5-deployment"
        }
      }
    }
  }
}
    '

Replace ROUTE_NAME|ID with the id or name of the route that this plugin configuration will target.

Make the following request, substituting your own access token, region, control plane ID, and route ID:

curl -X POST \
https://{us|eu}.api.konghq.com/v2/control-planes/{controlPlaneId}/core-entities/routes/{routeId}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --header "Authorization: Bearer TOKEN" \
    --data '{"name":"ai-response-transformer","config":{"prompt":"For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.","transformation_extract_pattern":"\\\\{((.|\\n)*)\\\\}","parse_llm_response_json_instructions":false,"llm":{"route_type":"llm/v1/chat","auth":{"header_name":"api-key","header_value":"<AZURE_OPENAI_TOKEN>"},"logging":{"log_statistics":true,"log_payloads":false},"model":{"provider":"azure","name":"gpt-35-turbo","options":{"max_tokens":1024,"temperature":1.0,"azure_instance":"azure-openai-instance-name","azure_deployment_id":"gpt-3-5-deployment"}}}}}'

See the Konnect API reference to learn about region-specific URLs and personal access tokens.

First, create a KongPlugin resource:

echo "
apiVersion: configuration.konghq.com/v1
kind: KongPlugin
metadata:
  name: ai-response-transformer-example
plugin: ai-response-transformer
config:
  prompt: For any city name, put the country that it's in, in brackets next to it. Reply
    with only the JSON result.
  transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
  parse_llm_response_json_instructions: false
  llm:
    route_type: llm/v1/chat
    auth:
      header_name: api-key
      header_value: "<AZURE_OPENAI_TOKEN>"
    logging:
      log_statistics: true
      log_payloads: false
    model:
      provider: azure
      name: gpt-35-turbo
      options:
        max_tokens: 1024
        temperature: 1.0
        azure_instance: azure-openai-instance-name
        azure_deployment_id: gpt-3-5-deployment
" | kubectl apply -f -

Next, apply the KongPlugin resource to an ingress by annotating the ingress as follows:

kubectl annotate ingress INGRESS_NAME konghq.com/plugins=ai-response-transformer-example

Replace INGRESS_NAME with the name of the ingress that this plugin configuration will target. You can see your available ingresses by running kubectl get ingress.

Note: The KongPlugin resource only needs to be defined once and can be applied to any service, consumer, or route in the namespace. If you want the plugin to be available cluster-wide, create the resource as a KongClusterPlugin instead of KongPlugin.

Add this section to your declarative configuration file:

plugins:
- name: ai-response-transformer
  route: ROUTE_NAME|ID
  config:
    prompt: For any city name, put the country that it's in, in brackets next to it. Reply
      with only the JSON result.
    transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
    parse_llm_response_json_instructions: false
    llm:
      route_type: llm/v1/chat
      auth:
        header_name: api-key
        header_value: "<AZURE_OPENAI_TOKEN>"
      logging:
        log_statistics: true
        log_payloads: false
      model:
        provider: azure
        name: gpt-35-turbo
        options:
          max_tokens: 1024
          temperature: 1.0
          azure_instance: azure-openai-instance-name
          azure_deployment_id: gpt-3-5-deployment

Replace ROUTE_NAME|ID with the id or name of the route that this plugin configuration will target.

Prerequisite: Configure your Personal Access Token 
terraform {
  required_providers {
    konnect = {
      source  = "kong/konnect"
    }
  }
}

provider "konnect" {
  personal_access_token = "kpat_YOUR_TOKEN"
  server_url            = "https://us.api.konghq.com/"
}

Add the following to your Terraform configuration to create a Konnect Gateway Plugin:

resource "konnect_gateway_plugin_ai_response_transformer" "my_ai_response_transformer" {
  enabled = true

  config = {
    prompt = "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result."
    transformation_extract_pattern = "\\{((.|\n)*)\\}"
    parse_llm_response_json_instructions = false
    llm = {
      route_type = "llm/v1/chat"
      auth = {
        header_name = "api-key"
        header_value = "<AZURE_OPENAI_TOKEN>"
      }
      logging = {
        log_statistics = true
        log_payloads = false
      }
      model = {
        provider = "azure"
        name = "gpt-35-turbo"
        options = {
          max_tokens = 1024
          temperature = 1.0
          azure_instance = "azure-openai-instance-name"
          azure_deployment_id = "gpt-3-5-deployment"
        }
      }
    }
  }

  control_plane_id = konnect_gateway_control_plane.my_konnect_cp.id
  route = {
    id = konnect_gateway_route.my_route.id
  }
}

The following examples provide some typical configurations for enabling the ai-response-transformer plugin on a consumer.

Make the following request:

curl -X POST http://localhost:8001/consumers/{consumerName|Id}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --data '
    {
  "name": "ai-response-transformer",
  "config": {
    "prompt": "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.",
    "transformation_extract_pattern": "\\\\{((.|\\n)*)\\\\}",
    "parse_llm_response_json_instructions": false,
    "llm": {
      "route_type": "llm/v1/chat",
      "auth": {
        "header_name": "api-key",
        "header_value": "<AZURE_OPENAI_TOKEN>"
      },
      "logging": {
        "log_statistics": true,
        "log_payloads": false
      },
      "model": {
        "provider": "azure",
        "name": "gpt-35-turbo",
        "options": {
          "max_tokens": 1024,
          "temperature": 1.0,
          "azure_instance": "azure-openai-instance-name",
          "azure_deployment_id": "gpt-3-5-deployment"
        }
      }
    }
  }
}
    '

Replace CONSUMER_NAME|ID with the id or name of the consumer that this plugin configuration will target.

Make the following request, substituting your own access token, region, control plane ID, and consumer ID:

curl -X POST \
https://{us|eu}.api.konghq.com/v2/control-planes/{controlPlaneId}/core-entities/consumers/{consumerId}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --header "Authorization: Bearer TOKEN" \
    --data '{"name":"ai-response-transformer","config":{"prompt":"For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.","transformation_extract_pattern":"\\\\{((.|\\n)*)\\\\}","parse_llm_response_json_instructions":false,"llm":{"route_type":"llm/v1/chat","auth":{"header_name":"api-key","header_value":"<AZURE_OPENAI_TOKEN>"},"logging":{"log_statistics":true,"log_payloads":false},"model":{"provider":"azure","name":"gpt-35-turbo","options":{"max_tokens":1024,"temperature":1.0,"azure_instance":"azure-openai-instance-name","azure_deployment_id":"gpt-3-5-deployment"}}}}}'

See the Konnect API reference to learn about region-specific URLs and personal access tokens.

First, create a KongPlugin resource:

echo "
apiVersion: configuration.konghq.com/v1
kind: KongPlugin
metadata:
  name: ai-response-transformer-example
plugin: ai-response-transformer
config:
  prompt: For any city name, put the country that it's in, in brackets next to it. Reply
    with only the JSON result.
  transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
  parse_llm_response_json_instructions: false
  llm:
    route_type: llm/v1/chat
    auth:
      header_name: api-key
      header_value: "<AZURE_OPENAI_TOKEN>"
    logging:
      log_statistics: true
      log_payloads: false
    model:
      provider: azure
      name: gpt-35-turbo
      options:
        max_tokens: 1024
        temperature: 1.0
        azure_instance: azure-openai-instance-name
        azure_deployment_id: gpt-3-5-deployment
" | kubectl apply -f -

Next, apply the KongPlugin resource to an ingress by annotating the KongConsumer object as follows:

kubectl annotate KongConsumer CONSUMER_NAME konghq.com/plugins=ai-response-transformer-example

Replace CONSUMER_NAME with the name of the consumer that this plugin configuration will target. You can see your available consumers by running kubectl get KongConsumer.

To learn more about KongConsumer objects, see Provisioning Consumers and Credentials.

Note: The KongPlugin resource only needs to be defined once and can be applied to any service, consumer, or route in the namespace. If you want the plugin to be available cluster-wide, create the resource as a KongClusterPlugin instead of KongPlugin.

Add this section to your declarative configuration file:

plugins:
- name: ai-response-transformer
  consumer: CONSUMER_NAME|ID
  config:
    prompt: For any city name, put the country that it's in, in brackets next to it. Reply
      with only the JSON result.
    transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
    parse_llm_response_json_instructions: false
    llm:
      route_type: llm/v1/chat
      auth:
        header_name: api-key
        header_value: "<AZURE_OPENAI_TOKEN>"
      logging:
        log_statistics: true
        log_payloads: false
      model:
        provider: azure
        name: gpt-35-turbo
        options:
          max_tokens: 1024
          temperature: 1.0
          azure_instance: azure-openai-instance-name
          azure_deployment_id: gpt-3-5-deployment

Replace CONSUMER_NAME|ID with the id or name of the consumer that this plugin configuration will target.

Prerequisite: Configure your Personal Access Token 
terraform {
  required_providers {
    konnect = {
      source  = "kong/konnect"
    }
  }
}

provider "konnect" {
  personal_access_token = "kpat_YOUR_TOKEN"
  server_url            = "https://us.api.konghq.com/"
}

Add the following to your Terraform configuration to create a Konnect Gateway Plugin:

resource "konnect_gateway_plugin_ai_response_transformer" "my_ai_response_transformer" {
  enabled = true

  config = {
    prompt = "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result."
    transformation_extract_pattern = "\\{((.|\n)*)\\}"
    parse_llm_response_json_instructions = false
    llm = {
      route_type = "llm/v1/chat"
      auth = {
        header_name = "api-key"
        header_value = "<AZURE_OPENAI_TOKEN>"
      }
      logging = {
        log_statistics = true
        log_payloads = false
      }
      model = {
        provider = "azure"
        name = "gpt-35-turbo"
        options = {
          max_tokens = 1024
          temperature = 1.0
          azure_instance = "azure-openai-instance-name"
          azure_deployment_id = "gpt-3-5-deployment"
        }
      }
    }
  }

  control_plane_id = konnect_gateway_control_plane.my_konnect_cp.id
  consumer = {
    id = konnect_gateway_consumer.my_consumer.id
  }
}

The following examples provide some typical configurations for enabling the ai-response-transformer plugin on a consumer group.

Make the following request:

curl -X POST http://localhost:8001/consumer_groups/{consumerGroupName|Id}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --data '
    {
  "name": "ai-response-transformer",
  "config": {
    "prompt": "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.",
    "transformation_extract_pattern": "\\\\{((.|\\n)*)\\\\}",
    "parse_llm_response_json_instructions": false,
    "llm": {
      "route_type": "llm/v1/chat",
      "auth": {
        "header_name": "api-key",
        "header_value": "<AZURE_OPENAI_TOKEN>"
      },
      "logging": {
        "log_statistics": true,
        "log_payloads": false
      },
      "model": {
        "provider": "azure",
        "name": "gpt-35-turbo",
        "options": {
          "max_tokens": 1024,
          "temperature": 1.0,
          "azure_instance": "azure-openai-instance-name",
          "azure_deployment_id": "gpt-3-5-deployment"
        }
      }
    }
  }
}
    '

Replace CONSUMER_GROUP_NAME|ID with the id or name of the consumer group that this plugin configuration will target.

Make the following request, substituting your own access token, region, control plane ID, and consumer group ID:

curl -X POST \
https://{us|eu}.api.konghq.com/v2/control-planes/{controlPlaneId}/core-entities/consumer_groups/{consumerGroupId}/plugins \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --header "Authorization: Bearer TOKEN" \
    --data '{"name":"ai-response-transformer","config":{"prompt":"For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.","transformation_extract_pattern":"\\\\{((.|\\n)*)\\\\}","parse_llm_response_json_instructions":false,"llm":{"route_type":"llm/v1/chat","auth":{"header_name":"api-key","header_value":"<AZURE_OPENAI_TOKEN>"},"logging":{"log_statistics":true,"log_payloads":false},"model":{"provider":"azure","name":"gpt-35-turbo","options":{"max_tokens":1024,"temperature":1.0,"azure_instance":"azure-openai-instance-name","azure_deployment_id":"gpt-3-5-deployment"}}}}}'

See the Konnect API reference to learn about region-specific URLs and personal access tokens.

First, create a KongPlugin resource:

echo "
apiVersion: configuration.konghq.com/v1
kind: KongPlugin
metadata:
  name: ai-response-transformer-example
plugin: ai-response-transformer
config:
  prompt: For any city name, put the country that it's in, in brackets next to it. Reply
    with only the JSON result.
  transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
  parse_llm_response_json_instructions: false
  llm:
    route_type: llm/v1/chat
    auth:
      header_name: api-key
      header_value: "<AZURE_OPENAI_TOKEN>"
    logging:
      log_statistics: true
      log_payloads: false
    model:
      provider: azure
      name: gpt-35-turbo
      options:
        max_tokens: 1024
        temperature: 1.0
        azure_instance: azure-openai-instance-name
        azure_deployment_id: gpt-3-5-deployment
" | kubectl apply -f -

Next, apply the KongPlugin resource to an ingress by annotating the KongConsumerGroup object as follows:

kubectl annotate KongConsumerGroup CONSUMER_GROUP_NAME konghq.com/plugins=ai-response-transformer-example

Replace CONSUMER_GROUP_NAME with the name of the consumer group that this plugin configuration will target. You can see your available consumer groups by running kubectl get KongConsumerGroup.

Note: The KongPlugin resource only needs to be defined once and can be applied to any service, consumer, consumer group, or route in the namespace. If you want the plugin to be available cluster-wide, create the resource as a KongClusterPlugin instead of KongPlugin.

Add this section to your declarative configuration file:

plugins:
- name: ai-response-transformer
  consumer_group: CONSUMER_GROUP_NAME|ID
  config:
    prompt: For any city name, put the country that it's in, in brackets next to it. Reply
      with only the JSON result.
    transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
    parse_llm_response_json_instructions: false
    llm:
      route_type: llm/v1/chat
      auth:
        header_name: api-key
        header_value: "<AZURE_OPENAI_TOKEN>"
      logging:
        log_statistics: true
        log_payloads: false
      model:
        provider: azure
        name: gpt-35-turbo
        options:
          max_tokens: 1024
          temperature: 1.0
          azure_instance: azure-openai-instance-name
          azure_deployment_id: gpt-3-5-deployment

Replace CONSUMER_GROUP_NAME|ID with the id or name of the consumer group that this plugin configuration will target.

Prerequisite: Configure your Personal Access Token 
terraform {
  required_providers {
    konnect = {
      source  = "kong/konnect"
    }
  }
}

provider "konnect" {
  personal_access_token = "kpat_YOUR_TOKEN"
  server_url            = "https://us.api.konghq.com/"
}

Add the following to your Terraform configuration to create a Konnect Gateway Plugin:

resource "konnect_gateway_plugin_ai_response_transformer" "my_ai_response_transformer" {
  enabled = true

  config = {
    prompt = "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result."
    transformation_extract_pattern = "\\{((.|\n)*)\\}"
    parse_llm_response_json_instructions = false
    llm = {
      route_type = "llm/v1/chat"
      auth = {
        header_name = "api-key"
        header_value = "<AZURE_OPENAI_TOKEN>"
      }
      logging = {
        log_statistics = true
        log_payloads = false
      }
      model = {
        provider = "azure"
        name = "gpt-35-turbo"
        options = {
          max_tokens = 1024
          temperature = 1.0
          azure_instance = "azure-openai-instance-name"
          azure_deployment_id = "gpt-3-5-deployment"
        }
      }
    }
  }

  control_plane_id = konnect_gateway_control_plane.my_konnect_cp.id
  consumer_group = {
    id = konnect_gateway_consumer_group.my_consumer_group.id
  }
}

A plugin which is not associated to any service, route, consumer, or consumer group is considered global, and will be run on every request.

  • In self-managed Kong Gateway Enterprise, the plugin applies to every entity in a given workspace.
  • In self-managed Kong Gateway (OSS), the plugin applies to your entire environment.
  • In Konnect, the plugin applies to every entity in a given control plane.

Read the Plugin Reference and the Plugin Precedence sections for more information.

The following examples provide some typical configurations for enabling the AI Response Transformer plugin globally.

Make the following request:

curl -X POST http://localhost:8001/plugins/ \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --data '
    {
  "name": "ai-response-transformer",
  "config": {
    "prompt": "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.",
    "transformation_extract_pattern": "\\\\{((.|\\n)*)\\\\}",
    "parse_llm_response_json_instructions": false,
    "llm": {
      "route_type": "llm/v1/chat",
      "auth": {
        "header_name": "api-key",
        "header_value": "<AZURE_OPENAI_TOKEN>"
      },
      "logging": {
        "log_statistics": true,
        "log_payloads": false
      },
      "model": {
        "provider": "azure",
        "name": "gpt-35-turbo",
        "options": {
          "max_tokens": 1024,
          "temperature": 1.0,
          "azure_instance": "azure-openai-instance-name",
          "azure_deployment_id": "gpt-3-5-deployment"
        }
      }
    }
  }
}
    '

Make the following request, substituting your own access token, region, and control plane ID:

curl -X POST \
https://{us|eu}.api.konghq.com/v2/control-planes/{controlPlaneId}/core-entities/plugins/ \
    --header "accept: application/json" \
    --header "Content-Type: application/json" \
    --header "Authorization: Bearer TOKEN" \
    --data '{"name":"ai-response-transformer","config":{"prompt":"For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result.","transformation_extract_pattern":"\\\\{((.|\\n)*)\\\\}","parse_llm_response_json_instructions":false,"llm":{"route_type":"llm/v1/chat","auth":{"header_name":"api-key","header_value":"<AZURE_OPENAI_TOKEN>"},"logging":{"log_statistics":true,"log_payloads":false},"model":{"provider":"azure","name":"gpt-35-turbo","options":{"max_tokens":1024,"temperature":1.0,"azure_instance":"azure-openai-instance-name","azure_deployment_id":"gpt-3-5-deployment"}}}}}'

See the Konnect API reference to learn about region-specific URLs and personal access tokens.

Create a KongClusterPlugin resource and label it as global:

apiVersion: configuration.konghq.com/v1
kind: KongClusterPlugin
metadata:
  name: <global-ai-response-transformer>
  annotations:
    kubernetes.io/ingress.class: kong
  labels:
    global: "true"
config:
  prompt: For any city name, put the country that it's in, in brackets next to it. Reply
    with only the JSON result.
  transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
  parse_llm_response_json_instructions: false
  llm:
    route_type: llm/v1/chat
    auth:
      header_name: api-key
      header_value: "<AZURE_OPENAI_TOKEN>"
    logging:
      log_statistics: true
      log_payloads: false
    model:
      provider: azure
      name: gpt-35-turbo
      options:
        max_tokens: 1024
        temperature: 1.0
        azure_instance: azure-openai-instance-name
        azure_deployment_id: gpt-3-5-deployment
plugin: ai-response-transformer

Add a plugins entry in the declarative configuration file:

plugins:
- name: ai-response-transformer
  config:
    prompt: For any city name, put the country that it's in, in brackets next to it. Reply
      with only the JSON result.
    transformation_extract_pattern: "\\\\{((.|\\n)*)\\\\}"
    parse_llm_response_json_instructions: false
    llm:
      route_type: llm/v1/chat
      auth:
        header_name: api-key
        header_value: "<AZURE_OPENAI_TOKEN>"
      logging:
        log_statistics: true
        log_payloads: false
      model:
        provider: azure
        name: gpt-35-turbo
        options:
          max_tokens: 1024
          temperature: 1.0
          azure_instance: azure-openai-instance-name
          azure_deployment_id: gpt-3-5-deployment
Prerequisite: Configure your Personal Access Token 
terraform {
  required_providers {
    konnect = {
      source  = "kong/konnect"
    }
  }
}

provider "konnect" {
  personal_access_token = "kpat_YOUR_TOKEN"
  server_url            = "https://us.api.konghq.com/"
}

Add the following to your Terraform configuration to create a Konnect Gateway Plugin:

resource "konnect_gateway_plugin_ai_response_transformer" "my_ai_response_transformer" {
  enabled = true

  config = {
    prompt = "For any city name, put the country that it's in, in brackets next to it. Reply with only the JSON result."
    transformation_extract_pattern = "\\{((.|\n)*)\\}"
    parse_llm_response_json_instructions = false
    llm = {
      route_type = "llm/v1/chat"
      auth = {
        header_name = "api-key"
        header_value = "<AZURE_OPENAI_TOKEN>"
      }
      logging = {
        log_statistics = true
        log_payloads = false
      }
      model = {
        provider = "azure"
        name = "gpt-35-turbo"
        options = {
          max_tokens = 1024
          temperature = 1.0
          azure_instance = "azure-openai-instance-name"
          azure_deployment_id = "gpt-3-5-deployment"
        }
      }
    }
  }

  control_plane_id = konnect_gateway_control_plane.my_konnect_cp.id
}