Query OpenAI for analysis of data.
The Prompt parameter can be the entirety of the prompt to OpenAI, or variable data that is provided to support a question.
The PromptPrefix parameter can be used to preface data with a particular question or statement.
For example, consider the following:
Prompt = $DataBlob
PromptPrefix = What is your analysis of the data?
PromptPrefix + Prompt = What is your analysis of the data? $DataBlob
This artifact can be called from within another artifact (such as one looking for files) to enrich the data made available by that artifact.
Ex.
SELECT * from Artifact.Server.Enrichment.OpenAI(Prompt=$YOURPROMPT, PromptPrefix=$YOURPROMPTPREFIX)
name: Server.Enrichment.OpenAI
author: Wes Lambert - @therealwlambert|@weslambert@infosec.exchange
description: |
Query OpenAI for analysis of data.
The `Prompt` parameter can be the entirety of the prompt to OpenAI, or variable data that is provided to support a question.
The `PromptPrefix` parameter can be used to preface data with a particular question or statement.
For example, consider the following:
`Prompt` = $DataBlob
`PromptPrefix` = What is your analysis of the data?
`PromptPrefix` + `Prompt` = What is your analysis of the data? $DataBlob
This artifact can be called from within another artifact (such as one looking for files) to enrich the data made available by that artifact.
Ex.
`SELECT * from Artifact.Server.Enrichment.OpenAI(Prompt=$YOURPROMPT, PromptPrefix=$YOURPROMPTPREFIX)`
type: SERVER
parameters:
- name: Prompt
type: string
description: The data sent to OpenAI
default: 'This is a test'
- name: PromptPrefix
type: string
description: A prefix to be used with the prompt. For example, when asking a question, then providing data separately
default: ''
- name: MaxTokens
type: int
description: The maximum number of tokens returned in the completion
default: 100
- name: Model
type: string
description: The model used for processing the prompt
default: gpt-3.5-turbo
- name: OpenAIToken
type: string
description: Token for OpenAI. Leave blank here if using server metadata store.
default:
- name: Role
type: string
description: Role used for messages
default: system
- name: User
type: string
description: User for API request
default: abcd12345
sources:
- query: |
LET Creds <= if(
condition=OpenAIToken,
then=OpenAIToken,
else=server_metadata().OpenAIToken)
SELECT PromptPrefix + Prompt AS Prompt,
parse_json(data=Content).choices[0].message.content AS ResponseText,
parse_json(data=Content) AS ResponseDetails
FROM http_client(
url='https://api.openai.com/v1/chat/completions',
headers=dict(`Authorization`='Bearer ' + Creds, `Content-Type`="application/json"),
method="POST",
data=serialize(item=dict(`model`=Model, messages=[dict(`role`=Role, `content`=Prompt), ]))
)