> ## Documentation Index
> Fetch the complete documentation index at: https://nenyax.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Call Intelligence

> AI-generated analysis of every completed call.

# Call Intelligence

After every completed call, Nenyax automatically analyzes the conversation transcript and generates structured insights.

## What's Analyzed

| Field               | Description                                                   |
| ------------------- | ------------------------------------------------------------- |
| **Summary**         | 2-3 sentence overview of the conversation                     |
| **Sentiment**       | Overall caller sentiment: positive, neutral, or negative      |
| **Sentiment Score** | Confidence score (0-1) for the sentiment classification       |
| **Outcome**         | Call outcome: resolved, unresolved, transferred, or escalated |
| **Topics**          | Key topics discussed during the call                          |
| **Action Items**    | Follow-up actions identified from the conversation            |

## How It Works

1. When a session ends (`POST /api/sessions/by-room/{room_name}/end`), the backend collects all transcripts
2. The full conversation is sent to the AI analysis service
3. The analysis result is stored in the session's `analysis` JSON column
4. Results are displayed on the call details page

## Viewing Analysis

Navigate to **Dashboard → Call Logs → \[Session]** to see the analysis card with:

* Summary text
* Sentiment badge with confidence percentage
* Outcome badge (color-coded)
* Topic tags
* Action item checklist

## API Access

```bash theme={null}
# Get analysis for a session
curl http://localhost:8000/api/sessions/{session_id}/analysis
```

Response:

```json theme={null}
{
  "summary": "Customer called about a billing discrepancy on their latest invoice...",
  "sentiment": "neutral",
  "sentiment_score": 0.65,
  "outcome": "resolved",
  "topics": ["billing", "invoice", "refund"],
  "action_items": [
    "Process refund for $45.00",
    "Send updated invoice to customer email"
  ]
}
```