Quickstart

Prerequisites

• Python 3.11+
• A workspace API key (open the dashboard)
• A configured API origin for your workspace

Install the CLI

python3 -m pip install schift-cli
export SCHIFT_API_KEY=sch_...
export SCHIFT_API_URL=https://api.example.com/v1

Create a searchable bucket

Start with one document. That proves your key, API origin, upload path, and search path before you build a UI around it.

schift db create support-docs
schift upload ./handbook.pdf --bucket support-docs
schift search "How do I reset my password?" --bucket support-docs --top-k 5

The search command returns the best matching chunks and source metadata.

Add the same bucket to an agent

Use the TypeScript SDK when you are ready to embed the retrieval path in an app or service.

import { WorkspaceClient, Agent, RAG } from "@schift-io/sdk";

const client = new WorkspaceClient({ apiKey: process.env.SCHIFT_API_KEY });
const rag = new RAG({ bucket: "support-docs" }, client.transport);

const agent = new Agent({
  name: "Support Bot",
  instructions: "Answer questions using the knowledge base.",
  rag,
  model: "gpt-4o-mini",
  transport: client.transport,
});

• WorkspaceClient connects to the configured API origin
• RAG wraps a document bucket (OCR, chunking, embedding, search — all managed)
• Agent runs a ReAct loop: receive question -> search docs -> generate answer

App starters

Start a new project from the SDK shape above. For terminal workflows, use the Python schift-cli.

Next steps

• RAG — upload documents and configure search