Unajengaje plugin au programu yenye AI ndani, kwa mfano kuunganisha Claude API katika plugin ya WordPress?

Question

Accepted Answer

Unajumuisha AI katika programu kwa kuita API ya modeli **upande wa severi** na kutibu pato lake kama ingizo ambayo haiaminiki. Kwa Claude, hii inamaanisha kuita **Messages API** kutoka kwenye backend yako (PHP, kwa plugin ya WordPress) na ufunguo wa API uhifadhiwe kwenye severi — kamwe si katika JavaScript upande wa mteja, ambapo mtu yeyote anaweza kuiba.

## Simu upande wa severi

Kanuni kuu: **ufunguo wa API unaishi kwenye severi**. Kivinjari kinazungumza na *ncha yako* ya mwisho; *ncha yako* ya mwisho inazungumza na 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
```

## Zaidi ya misingi

```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
```

Kila wakati **thibitisha pato la modeli kabla ya kumtumia**: usiitendee kama nambari, usiitaje bila kutoroka, au usiandike kwenye DB yako bila kugeuza. Modeli ni jenereta ya maandishi yenye nguvu lakini inayoweza kosa, si chanzo kinachoadhaniwa.

## Kwa nini ni muhimu

Kusambaza AI ndani ya bidhaa halisi ni hasa uhandisi wa kuchosha lakini muhimu karibu na simu, si simu yenyewe. Kuweka ufunguo kwenye severi kunazuia ujambazi na bili zisizobadilika; kumudu makosa, mipaka ya kasi, na gharama huweka tabia kuwa na kutegemewa na inakabiliwa; na kugeuza pato kabla ya kumtumia kunazuia makosa ya modeli au amri iliyoingizwa kuwa shimo la usalama. Ipata sahihi na Messages API inakuwa jivu la kujenga kuaminika; ruka hizo na mademo ya kushangilia hubadilika kuwa ufunguo ulioibwa, bili ya heri, au ubogaji.