ZephyrCode JavaScript SDK

Official JavaScript SDK for ZephyrCode — the AI coding agent platform

Installation

npm i zephyrcode

Note: This is a closed-source, proprietary SDK. There is no public repository. For support, email [email protected].

Quickstart

1. Get an API key

Create one at zephyrcode.space-z.ai/apikey. Your key will look like:

zephyr-yourname-aBcD123eFgH456...

2. Stream a chat completion

const { ZephyrCode } = require("zephyrcode");

const client = new ZephyrCode({ apiKey: "zephyr-yourname-xxx" });

(async () => {
  for await (const chunk of client.chat.stream({
    message: "Build a React login form with Tailwind CSS",
    model: "z-code-ultra",
    mode: "code",
  })) {
    if (chunk.type === "content") {
      process.stdout.write(chunk.delta);
    } else if (chunk.type === "done") {
      console.log(`\n\n✅ Done — tools used: ${chunk.toolsUsed}`);
    }
  }
})();

3. Non-streaming response

const response = await client.chat.create({
  message: "Explain async/await in JavaScript",
  model: "z-code-pro",
});
console.log(response.content);
console.log(`Reasoning steps: ${response.reasoning.length}`);
console.log(`Tool calls: ${response.toolCalls.length}`);

API Reference

new ZephyrCode({ apiKey, baseUrl, timeout })

| Parameter | Type | Default | Description | |-----------|------|---------|-------------| | apiKey | string | required | Your zephyr-{username}-{secret} key | | baseUrl | string | https://zephyrcode.space-z.ai | API base URL | | timeout | number | 60000 | Request timeout in ms |

client.chat.stream(params) — AsyncIterable

Stream a chat completion as Server-Sent Events.

| Parameter | Type | Default | Description | |-----------|------|---------|-------------| | message | string | — | A single user message | | messages | Message[] | — | Full conversation history | | model | string | "z-code-ultra" | Model ID | | mode | string | "code" | Mode | | thinking | boolean | true | Enable extended reasoning | | projectContext | object | — | {language, framework, files} |

Yields event objects with .type:

| Type | Description | |------|-------------| | meta | Stream start — model + mode info | | plan | Task decomposition with steps | | reasoning | Chain-of-thought chunk | | tool_call | Agent invokes a tool | | tool_result | Structured tool output | | content | Markdown content delta (.delta) | | done | Terminal event with stats | | error | Fatal error (throws StreamError) |

client.chat.create(params) → Promise<ChatResponse>

Same parameters as .stream(), returns { content, model, mode, toolsUsed, fallbackUsed, reasoning, toolCalls, plan }.

client.tts.create({ text, voice, speed, translate, targetLanguage }) → Promise<TTSResponse>

const audio = await client.tts.create({
  text: "Hello, welcome to ZephyrCode!",
  voice: "rachel",
  speed: 1.0,
});
require("fs").writeFileSync("welcome.wav", audio.audio);

client.apikeys — API Key Management

const keys = await client.apikeys.list();
const newKey = await client.apikeys.create({
  name: "Production",
  permissions: "all",        // "all", "restricted", "readonly"
  expiration: "60d",         // "1d", "3d", "7d", "21d", "60d", "never"
  credit: 100.0,
  project: "My App",
});
await client.apikeys.revoke(newKey.id);
await client.apikeys.delete(newKey.id);

client.projects — Project Management

const projects = await client.projects.list();
const project = await client.projects.create("My App");
await client.projects.delete(project.id);

client.models — Model Information

const models = await client.models.list();
models.forEach((m) => console.log(`${m.id} — ${m.label}`));

Models

| Model | ID | Best for | |-------|----|----------| | Z Code Ultra | z-code-ultra | Frontier reasoning, multi-file refactors | | Z Code Pro | z-code-pro | Everyday coding, quick fixes | | Z Code Local | z-code-local | On-device, privacy-first |

Error Handling

const { ZephyrCode, AuthenticationError, RateLimitError, APIError, StreamError } = require("zephyrcode");

try {
  const response = await client.chat.create({ message: "Hello" });
} catch (e) {
  if (e instanceof AuthenticationError) {
    console.log("Invalid API key");
  } else if (e instanceof RateLimitError) {
    console.log(`Rate limited. Retry after ${e.retryAfter}s`);
  } else if (e instanceof APIError) {
    console.log(`API error (${e.statusCode}): ${e.message}`);
  } else if (e instanceof StreamError) {
    console.log(`Stream error: ${e.message}`);
  }
}

Automatic Retries

The SDK automatically retries on 502, 503, 504 status codes and connection errors — up to 3 times with exponential backoff (1s, 2s, 4s). This handles intermittent gateway errors transparently.

ESM Import

import { ZephyrCode } from "zephyrcode";

CommonJS Require

const { ZephyrCode } = require("zephyrcode");

License

Proprietary © ZephyrCode Labs. This SDK is closed-source. All rights reserved.

Links

• Documentation: zephyrcode.space-z.ai/docs
• API Keys: zephyrcode.space-z.ai/apikey
• Support: [email protected]