GET
/
api
/
external
/
export
/
interactions
/
detail-v2
/
{interaction_id}
curl --request GET \
  --url https://backend.nebuly.com/api/external/export/interactions/detail-v2/{interaction_id} \
  --header 'Authorization: Bearer <token>'
{
  "id": "<string>",
  "project_id": "<string>",
  "sentiment": "<string>",
  "emotions": {},
  "cost": 123,
  "topic": "<string>",
  "intent": "<string>",
  "input": "<string>",
  "output": "<string>",
  "latency": 123,
  "pii_removed": true,
  "text_language": "<string>",
  "timestamp": "<string>",
  "end_timestamp": "<string>",
  "user": "<string>",
  "verbosity": "<string>",
  "actions": [
    "<string>"
  ],
  "action_details": "<string>",
  "feedback_actions": [
    "<string>"
  ],
  "feedback_action_text": "<string>",
  "feedback_action_rating": 123,
  "user_query": "<string>",
  "tags": {},
  "feedback_type": "<string>",
  "risky_behavior": "<string>",
  "rags": [
    "<string>"
  ],
  "rags_details": {},
  "warning_message": "<string>",
  "conversation": [
    {
      "input": "<string>",
      "output": "<string>",
      "timestamp": "<string>",
      "is_chosen": true
    }
  ]
}

Retrieve the detailed information about a single interaction using the v2 schema. This version includes expanded metadata such as sentiment, emotions, latency, PII info, and feedback details. You can list interactions using the Get interactions endpoint. For authentication you need the to use the API key associated to the project you want to retrieve the interactions from.

​
interaction_id
UUID
required

ID of the interaction you want to retrieve.

​
Response

​
id
string

Unique identifier of the interaction.

​
project_id
string

The project identifier to which the interaction belongs.

​
sentiment
string

Sentiment classification of the interaction, if available (e.g., positive, neutral, negative).

​
emotions
object

Key-value pairs describing emotion probabilities or intensities detected in the interaction.

​
cost
number

The cost incurred for the interaction

​
topic
string

Automatically or manually assigned topic of the interaction.

​
intent
string

The detected intent associated with the interaction.

​
input
string

The user input as received by the system.

​
output
string

The LLM-generated output corresponding to the input.

​
latency
number

Time taken to generate the response.

​
pii_removed
boolean

Flag indicating whether PII (Personally Identifiable Information) was removed from the input.

​
text_language
string

Language detected in the user input.

​
timestamp
string

The timestamp when the interaction began (user input was received).

​
end_timestamp
string

The timestamp when the assistant’s response was generated.

​
user
string

User identifier associated with the interaction.

​
verbosity
string

Number of characters in the output.

​
actions
string[]

List of action labels inferred or triggered from the interaction

​
action_details
string

Additional context or metadata about the actions taken.

​
feedback_actions
string[]

Actions taken or suggested via user feedback (e.g., “thumbs_up”, “thumbs_down”).

​
feedback_action_text
string

Text content accompanying feedback actions, if any.

​
feedback_action_rating
number

Numerical rating provided as feedback (e.g., 1–5).

​
user_query
string

A normalized or cleaned-up version of the original user query, if processed.

​
tags
object

Key-value pairs showing custom tags associated with the interaction.

​
feedback_type
string

Type of feedback provided (e.g., “negative_implicit”, “positive_implicit”).

​
risky_behavior
string

A flag or description for any risky or concerning user behavior detected.

​
rags
string[]

Identifiers or summaries of RAG sources used, if provided.

​
rags_details
object

Detailed metadata about each RAG item (e.g., source, query, result).

​
warning_message
string

Description of any warnings or issues detected with the LLM response.

​
conversation
object[]

List of conversation turns surrounding the interaction.