Create Profile

Version: Beta

Create the summarization Profile that specifies both prompting instructions and session context configuration.

Request Body required

slice object optional

mode remove earliest or latest events as defined optional

Possible values: [UNSPECIFIED, FIRST, LAST, TIMESTAMP]

Default value: FIRST

event_limit int32 optional

duration_limit_ms int64 optional

start_timestamp date-time optional

If specified, the event limit and duration limit will be applied to the time beginning at this timestamp.

context object optional

include string[] optional

Controls the session context elements to include, takes precedence over individual inclusion fields. Can be 'user', 'org', 'location', 'device'. If not specified and not excluded, all context elements are included.

exclude string[] optional

Controls the session context elements to exclude, takes precedence over individual exclusion fields. Can be 'user', 'org', 'location', 'device'. If not specified and not excluded, all context elements are included.

exclude_org_context boolean optional

exclude_user_context boolean optional

exclude_location boolean optional

exclude_device boolean optional

exclude_descriptions boolean optional

events object optional

exclude_types string[] optional

include_types string[] optional

exclude_defined_events boolean optional

exclude_api_events boolean optional

exclude_event_timestamps boolean optional

exclude_selectors boolean optional

include_selector_tags boolean optional

trim_to_last_n_selectors int32 optional

cache object optional

enable_event_cache boolean optional

llm object optional

Configuration used to select the LLM model and related prompt and inference options.

pre_prompt string optional

Text to be included in the Generative AI prompt before including the contextual representation of the requested session.

post_prompt string optional

Text to be included in the Generative AI prompt before including the contextual representation of the requested session.

output_schema Optional JSON Schema to define output optional

JSON schema to provide to Generative AI for formatting of result and to be used to validate the proper output format.

model string optional

Possible values: [GEMINI_2_FLASH, GEMINI_2_FLASH_LITE]

Default value: GEMINI_2_FLASH_LITE

Generative AI large language model to use.

temperature float optional

Controls the randomness of text generated by a large language model.

Responses

200

A successful response.

Schema

Example

{}

Returned when invalid input has been provided. Fix the issue and retry.

Schema

Example

{
  "message": "session_id is required",
  "code": "required_field"
}

Returned when access to the resource is unauthorized.

Schema

Example

{
  "message": "access is unauthorized",
  "code": "unauthorized"
}

Returned when access is not allowed due to insufficient permissions.

Schema

Example

{
  "message": "insufficient permissions",
  "code": "forbidden"
}

Returned when the resource does not exist.

Schema

Example

{
  "message": "Requested resource does not exist",
  "code": "resource_not_found"
}

Returned when the client has exceeded the rate limit for this endpoint. A Retry-After header will be included with the response. This header will contain the number of seconds that you should wait before attempting to send another request.

Schema

Example

{
  "message": "Too many requests. Client has exceeded the rate limit for this endpoint.",
  "code": "too_many_requests"
}

Returned when a server error is encountered.

Schema

Example

{
  "message": "Server error was encountered",
  "code": "server_error"
}

Create Profile​

Create Profile