Paano ka nag-build ing plugin o app na may AI sa loob, contoh integrating ing Claude API sa WordPress plugin?

Question

Accepted Answer

Sampeyan build AI sa app kanthi tawon API model **server-side** lan memperlakukan output minangka untrusted input. Kanggo Claude, iku tegese tawon **Messages API** saka backend sampeyan (PHP, kanggo plugin WordPress) karo API key disimpen ing server — ora ing client-side JavaScript, nandhang sapa wae bisa nyolong.

## Apa iku penting

Kaidah utama: **API key nduwur ing server**. Browser omong karo *endpoint* sampeyan; *endpoint* sampeyan omong karo Claude.

```php
<?php
// In a WordPress plugin: run this in PHP (server-side), never expose the key to JS.
$resp = wp_remote_post( 'https://api.anthropic.com/v1/messages', [
  'headers' => [
    'x-api-key'         => getenv( 'ANTHROPIC_API_KEY' ), // from env/secret store, not code
    'anthropic-version' => '2023-06-01',
    'content-type'      => 'application/json',
  ],
  'timeout' => 30,
  'body' => wp_json_encode( [
    'model'      => 'claude-sonnet-4-5',   // pick a current model id
    'max_tokens' => 1024,                   // cap output → controls cost
    'messages'   => [
      [ 'role' => 'user', 'content' => $user_prompt ],
    ],
  ] ),
] );

if ( is_wp_error( $resp ) ) { /* handle network/timeout errors, maybe retry */ }
$data = json_decode( wp_remote_retrieve_body( $resp ), true );
$text = $data['content'][0]['text'] ?? '';   // validate before trusting/displaying it
```

## Beyond basics

```text
- TOOL USE / FUNCTION CALLING → let the model call your functions (search, DB lookup)
- STREAMING                   → stream tokens for a responsive UI on long answers
- PROMPT CACHING              → cache a large static system prompt → cheaper, faster
- STRUCTURED OUTPUT           → ask for JSON, then parse + schema-validate it
- ERRORS & RATE LIMITS        → handle 429/5xx with backoff + retry; show a fallback
- COST                        → cap max_tokens, log token usage, set per-user limits
```

Salawiyé **validate model output sadurungé gawe**: ora bakal njalanaken minangka code, ora bakal render tanpa escaping, utawa ora bakal nulis ing DB sampeyan tanpa checked. Model iku generator teks kuwat nanging boten dipercaya, bukan sumber terpercaya.

## Apa iku penting

Gawe AI sa produk nyata mostly boring-nanging-kritis engineering sakulinan tawon, bukan tawon dhewe. Keeping key server-side ngegat theft lan runaway bills; nanggulangi errors, rate limits, lan cost njaga feature reliable lan affordable; lan validate output sadurungé acting on it ngegat mistakes model utawa injected instruction saka dadi security hole. Get sini tengen lan Messages API dadi dependable building block; skip sini lan impressive demo dadi leaked key, surprise invoice, utawa exploit.