onairos/DEVELOPER_UPDATES.md

The Onairos Library is a collection of functions that enable Applications to connect and communicate data with Onairos Identities via User Authorization. Integration for developers is seamless, simple and effective for all applications. LLM SDK capabiliti

# Onairos SDK - Developer Updates

This file contains concise update summaries for developers integrating the Onairos SDK.

**Rollup (Apr 24 – May 2, 2026, v8.0.0 → v8.1.9):** see **`docs/SDK_PARITY_ROLLUP_2026_04_24_to_05_02.md`** for a **spec / API parity** summary (data request, training, persona train, pnpm/exports, and a checklist for non-web SDKs). Use it when updating other language SDKs and public docs.

---

## v6.2.1 - Inference Result Formatting Options

**Release Date:** April 4, 2026

### What's New

You can now control how inference results are sorted, ranked, and filtered directly from the SDK.

### New Parameters

| Parameter | Type | Description |
|-----------|------|-------------|
| `sortResults` | `'desc'` \| `'asc'` \| `false` | Sort results by score (descending or ascending) |
| `includeRank` | `boolean` | Add rank numbers (1, 2, 3...) to each result |
| `topK` | `number` | Return only the top K results |
| `groupByScore` | `{ precision: number }` | Group items with the same score (precision = decimal places) |
| `inferenceResponseFormat` | `string` | Output structure (e.g., `'ranked'`, `'grouped'`) |
| `includeOriginalIndex` | `boolean` | Preserve the original input index in results |

### Supported Endpoints

These parameters work with all inference endpoints:
- `/inferenceNoProof`
- `/mobileInferenceNoProof`
- `/combinedInference`
- `/combined-training-inference`

### Usage Examples

**Get top 5 highest-scoring items, sorted descending:**
```jsx
<OnairosButton
  webpageName="MyApp"
  inferenceData={eventTags}
  sortResults="desc"
  topK={5}
  includeRank={true}
  onComplete={(data) => console.log(data)}
/>
```

**Group items by score with 1 decimal precision:**
```jsx
<OnairosButton
  webpageName="MyApp"
  inferenceData={items}
  inferenceResponseFormat="grouped"
  groupByScore={{ precision: 1 }}
  onComplete={(data) => console.log(data)}
/>
```

### Response Metadata

When formatting options are used, the response includes `formattingApplied`:

```json
{
  "formattingApplied": {
    "sortResults": "desc",
    "topK": 5,
    "includeRank": true,
    "originalCount": 100,
    "returnedCount": 5
  }
}
```

### Backward Compatibility

✅ **No breaking changes** - Existing integrations work without modification.

### Update Instructions

```bash
npm install onairos@6.2.1
```

---

## Previous Updates

### v6.2.0 - Security Hardening & Native Packaging
- Pinned axios to 1.14.0 for security
- Added Capacitor native Google Sign-In support
- ChatGPT export fetches up to 30 conversations

### v5.8.23 - Cross-Site Session Detection
- Cross-site SSO for partner apps
- Seamless authentication across domains

### v5.8.22 - Session Persistence Fix
- Fixed cached login persistence
- Data Request UI improvements

---

*For full changelog, see [CHANGELOG.md](./CHANGELOG.md)*