PeakPrivacy Assistants API

Auth

HEADERS

Key	Value
Api-token	api_token from [https://app.peakprivacy.ch/api-tokens](https://app.peakprivacy.ch/api-tokens)
X-Requested-With	XMLHttpRequest
Accept	application/json

401 - Unauthenticated; API Token is invalid or expired OR Subscription is inactive

{
    "error": "Unauthenticated"
}

POST Create Assistant

https://api.peakprivacy.ch/v1/ai/assistants

REQUEST BODY SCHEMA: application/json

Key	Value
name	string (required) Assistant Name
model	string (required) Specifies the model which will be applied to the assistant by default Supported models : gpt-4-1106-preview gpt-4 gpt-3.5-turbo-1106 mistral-tiny mistral-small mistral-medium mistral-swiss
instructions	text (optional) Additional instructions for the assistant. Example: 'Please provide the answer in Spanish'
temperature	number (required) [ 0 .. 1 ] What sampling temperature to use, between 0.0 and 1.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
anonymize	boolean (required) Default: true Whether to use anonymization for the prompt. Might slightly increase response time due to anonymzation & deanonymization algorithms work.
file_ids	array of strings (optional) IDs of files to be attached to the assistant

REQUEST BODY EXAMPLE

{
   "name": "Invoices",
   "model": "mistral-swiss",
   "instructions": "Please provide the answer in Spanish",
  
  "temperature": 0.5,
  "anonymize": true,
  "file_ids": ["b7454f54-aa96-4306-b64d-83343d7bf4a4"]
}

Example request

curl --location --request POST 'https://api.peakprivacy.ch/v1/ai/assistants' \
--header 'Api-token: api-token' \
--header 'X-Requested-With: XMLHttpRequest' \
--header 'Accept: application/json' \
--header 'Content-Type:/ application/json' \
--data '{
   "name": "Invoices",
   "model": "mistral-swiss",
   "instructions": "Please provide the answer in Spanish",
  
  "temperature": 0.5,
  "anonymize": true,
  "file_ids": ["b7454f54-aa96-4306-b64d-83343d7bf4a4"]
}'

Possible Responses

201 - Success

{
    "id": "7ca6999d-3ed4-47d7-8915-4df8aab6c989",
    "name": "Invoices",
    "model": "mistral-swiss",
    "instructions": "Please provide the answer in Spanish",
    "temperature": 0.5,
    "anonymize": true,
    "files": [
        {
            "id": "b7454f54-aa96-4306-b64d-83343d7bf4a4",
            "name": "twint_pat_instructions_1.pdf",
            "ext": "pdf",
            "size": "2371319",
            "mimetype": "application/pdf"
        }
    ]
}

422 - Validation Error

{
    "message": "Validation error",
    "errors": {
        "model": [
            "The model field must be a string.",
            "The selected model is invalid."
        ],
        "name": [
            "The name field is required."
        ]
    }
}

PUT Update Assistant

https://api.peakprivacy.ch/v1/ai/assistants/{id}

REQUEST BODY SCHEMA: application/json

Key	Value
name	string Assistant Name
model	string Specifies the model which will be applied to the assistant by default Supported models : gpt-4-1106-preview gpt-4 gpt-3.5-turbo-1106 mistral-tiny mistral-small mistral-medium mistral-swiss
instructions	text Additional instructions for the assistant. Example: 'Please provide the answer in Spanish'
temperature	number [ 0 .. 1 ] What sampling temperature to use, between 0.0 and 1.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
anonymize	boolean Whether to use anonymization for the prompt. Might slightly increase response time due to anonymzation & deanonymization algorithms work.
file_ids	array of strings (optional) IDs of files to be attached to the assistant NOTE: update is performed with 'sync' logic, so if some of attached files are absent in the provided array, they will be detached of the assistant. Empty array will detach all files. To leave files list untouched don't send this parameter

REQUEST BODY EXAMPLE

{
   "name": "Assistant Name UPDATED",
   "model": "mistral-swiss",
   "instructions": "Please provide the answer in Spanish",
    "temperature": 0.7,
    "anonymize": true,
    "file_ids": ["b7454f54-aa96-4306-b64d-83343d7bf4a4"]
}

Example request

curl --location --request PUT 'https://api.peakprivacy.ch/v1/ai/assistants/d9ba620b-4b8b-4ded-89be-d93b08ad786d' \
--header 'Api-token: api-token' \
--header 'X-Requested-With: XMLHttpRequest' \
--header 'Accept: application/json' \
--header 'Content-Type: application/json' \
--data '{
   "name": "Assistant Name UPDATED",
   "model": "mistral-swiss",
   "instructions": "Please provide the answer in Spanish",
    "temperature": 0.7,
    "anonymize": true,
    "file_ids": ["b7454f54-aa96-4306-b64d-83343d7bf4a4"]
}'

Possible Responses

200 - Success

{
    "id": "d9ba620b-4b8b-4ded-89be-d93b08ad786d",
    "name": "Assistant Name UPDATED",
    "model": "mistral-swiss",
    "instructions": "Please provide the answer in Spanish",
    "temperature": 0.7,
    "anonymize": true,
    "files": [
        {
            "id": "b7454f54-aa96-4306-b64d-83343d7bf4a4",
            "name": "twint_pat_instructions_1.pdf",
            "ext": "pdf",
            "size": "2371319",
            "mimetype": "application/pdf"
        }
    ]
}

422 - Validation Error

{
    "message": "Validation error",
    "errors": {
        "model": [
            "The model field must be a string.",
            "The selected model is invalid."
        ],
        "name": [
            "The name field is required."
        ]
    }
}

404 - Not Found

{
    "error": "Not found"
}

GET Retrieve Assistant

https://api.peakprivacy.ch/v1/ai/assistants/{id}

Example request

curl --location --request GET 'https://api.peakprivacy.ch/v1/ai/assistants/c33a8b18-c82a-4955-a449-515001848b1f' \
--header 'Api-token: api-token' \
--header 'X-Requested-With: XMLHttpRequest' \
--header 'Accept: application/json'

Possible Responses

200 - Success

{
    "id": "c33a8b18-c82a-4955-a449-515001848b1f",
    "name": "Assistant Name_UPDATED",
    "model": "gpt-3.5-turbo-1106",
    "instructions": "context_UPDATED (PUT)",
    "temperature": 0.7,
    "anonymize": true,
    "files": [
        {
            "id": "b7454f54-aa96-4306-b64d-83343d7bf4a4",
            "name": "twint_pat_instructions_1.pdf",
            "ext": "pdf",
            "size": "2371319",
            "mimetype": "application/pdf"
        }
    ]
}

404 - Not Found

{
    "error": "Not found"
}

GET Assistants paginated list

https://api.peakprivacy.ch/v1/ai/assistants

GET Parameters

Key	Value
limit	int (optional) default: 20 Per page items
cursor	string (optional) Pagination cursor

Example request

curl --location --request GET 'https://api.peakprivacy.ch/v1/ai/assistants?limit=6' \
--header 'API-TOKEN: api-token' \
--header 'Accept: application/json' \
--header 'Content-Type: application/json'

Possible Responses

200 - Success

{
   "data": [
        {
            "uuid": "8147e92d-2b98-4719-8ba1-d6a3a968e4cf",
            "name": "Invoices",
            "model": null,
            "instructions": null,
            "files": []
        },
        {
            "uuid": "98369e03-2aaa-4810-89c6-d05157965738",
            "name": "Invoices",
            "model": null,
            "instructions": null,
            "files": [
                {
                    "id": "fc2930a6-8d3f-4d8f-bb2f-5f75cced1aa6",
                    "name": "test-2",
                    "ext": "txt",
                    "size": "12345",
                    "mimetype": "text/plain-text"
                }
            ]
        }
    ],
    "links": {
        "first": null,
        "last": null,
        "prev": null,
        "next": "{{ $frontmatter.global.apiUrl }}/v1/ai/assistants?cursor=eyJhc3Npc3RhbnRzLmlkIjoyLCJfcG9pbnRzVG9OZXh0SXRlbXMiOnRydWV9"
    },
    "meta": {
        "path": "{{ $frontmatter.global.apiUrl }}/v1/ai/assistants",
        "per_page": 2,
        "next_cursor": "eyJhc3Npc3RhbnRzLmlkIjoyLCJfcG9pbnRzVG9OZXh0SXRlbXMiOnRydWV9",
        "prev_cursor": null
    }
}

DELETE Delete Assistant

https://api.peakprivacy.ch/v1/ai/assistants/{id}

Example request

curl --location --request DELETE 'https://api.peakprivacy.ch/v1/ai/assistants/a4904bba-e854-4709-9e48-8cbc149fff35' \
--header 'Api-token: api_token' \
--header 'X-Requested-With: XMLHttpRequest' \
--header 'Accept: application/json' \

Possible Responses

204 - Success

NO CONTENT

404 - Not Found

{
    "error": "Not found"
}

POST Ask Assistant

https://api.peakprivacy.ch/v1/ai/assistants/{id}/ask

REQUEST BODY SCHEMA: application/json

Key	Value
messages	required Array of objects The prompt(s) to generate completions for, encoded as a list of dict with role and content.
Following params are optional and override corresponding Assistant params
model	string Specifies the version of the GPT model to be used. Supported models : gpt-4-1106-preview gpt-4 gpt-3.5-turbo-1106 mistral-tiny mistral-small mistral-medium mistral-swiss
temperature	number or null [ 0 .. 1 ] What sampling temperature to use, between 0.0 and 1.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
anonymize	boolean or null Whether to use anonymization for the prompt. Might slightly increase response time due to anonymzation & deanonymization algorithms work.

messages SCHEMA:

Key	Value
role	string system user assistant Specifies the LLM the prompt will be sent to
content	string

BODY raw

{
    "model": "mistral-swiss",
    "messages": [

        {
            "role": "user",
            "content": "Create a summary of all my documents"
        }
    ],
    "temperature": 0.2,
    "anonymize" : true
}

Example request

curl --location --request POST 'https://api.peakprivacy.ch/v1/ai/assistants/c33a8b18-c82a-4955-a449-515001848b1f/ask' \
--header 'API-TOKEN: api-token' \
--header 'Accept: application/json' \
--header 'Content-Type: application/json' \
--data '{
    "messages": [
        {
            "role": "user",
            "content": "create document summary"
        }
    ]
}'

Possible Responses

RESPONSE SCHEMA: application/json

Key	Value
id	string
object	string
created	integer
model	string
choices	Array of objects
usage	object

choices SCHEMA: application/json

Key	Value
finish_reason	string
index	string
message	object

message SCHEMA: application/json

Key	Value
role	string
content	string

usage SCHEMA: application/json

Key	Value
completion_tokens	integer
prompt_tokens	integer
total_tokens	integer

200 - Success

{
    "id": "2820e9eb-22e9-40c6-a2fb-07d7ee06453d",
    "object": "chat.completion",
    "created": 1710243743,
    "model": "mistral-swiss",
    "choices": [
        {
            "finish_reason": "stop",
            "index": 0,
            "message": {
                "content": "The document contains instructions for downloading and using the TWINT configurator app, including information on selecting the app environment, version, and build, onboarding the app, topping up credit, and updating the app. It also provides details for iOS and Android devices.",
                "role": "assistant"
            }
        }
    ],
    "usage": {
        "completion_tokens": 54,
        "prompt_tokens": 1196,
        "total_tokens": 1250
    }
}

422 - Unprocessable Entity; Validation Error

{
    "message": "The model field is required.",
    "errors": {
        "model": [
            "The selected model is invalid."
        ]
    }
}

404 - Assistant Not Found

{
    "error": "Not found"
}