One client, every Console resource — chat, agents, browsers, vectors, files, tracing. Calm types, streaming-first, OpenAI-compatible where it matters.
Where this fits
The SDK is the application-layer client. The Console is the platform. Most teams need both, but for different reasons.
Reach for the SDK when you are
Go to Console docs when you are
Install
npm install @cognipeer/console-sdkpnpm add @cognipeer/console-sdkyarn add @cognipeer/console-sdkFirst call
import { ConsoleClient } from '@cognipeer/console-sdk';
const client = new ConsoleClient({ apiKey: process.env.COGNIPEER_API_KEY! });
const stream = await client.chat.completions.create({
model: 'gpt-4o',
stream: true,
messages: [
{ role: 'system', content: 'You are a precise assistant.' },
{ role: 'user', content: 'Summarize the Q3 release notes.' },
],
});
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}Entry points
| You are… | Open this | What you get |
|---|---|---|
| Deciding how Console and the SDK divide responsibility | Working with Console | A clear split between platform ops and app integration |
| Wiring the client into a new service | Getting started | Install, auth, base URL, the first request |
| Mapping existing HTTP calls to SDK methods | Console API mapping | One-to-one method-to-endpoint reference |
| Following an end-to-end example | Examples | Chat, streaming, RAG, memory, tracing — runnable |
Help
Open a GitHub issue — the team triages from there. For platform questions (provider keys, tenants, billing) head to the Console docs instead.