A unified storage SDK for object and blob backends. One small, honest API. Web-standards I/O. An escape hatch when you need the native client.

npm install files-sdk

import { Files } from "files-sdk";
import { s3 } from "files-sdk/s3";

const files = new Files({
  adapter: s3({ bucket: "uploads" }),
});

await files.upload("avatars/abc.png", file, { contentType: "image/png" });
const got = await files.download("avatars/abc.png");

Swap the adapter import (files-sdk/r2, files-sdk/gcs, files-sdk/azure, …) and the rest of your code stays the same.

• One API across providers — upload, download, head, delete, copy, list, url, signedUploadUrl. The shape is the same on S3, GCS, Azure, Vercel Blob, the local filesystem, and consumer providers like Dropbox.
• Web-standard I/O — bodies are Blob, File, ReadableStream, Uint8Array, ArrayBuffer, or string. No provider-specific types leak into your code.
• Escape hatch — every adapter exposes its native client at files.raw, so provider-specific features are one property access away.
• Tree-shakeable — each adapter is a separate entry point. You only bundle what you import.

A growing catalog covering S3 and S3-compatible stores, the major cloud blob platforms, edge/serverless blob services, the local filesystem, and consumer file providers. See files-sdk.dev for the current list and per-adapter setup.

A growing set of subpaths wrap a configured Files instance as ready-made tools for popular AI SDKs — currently the Vercel AI SDK (files-sdk/ai-sdk), OpenAI's Responses API and Agents SDK (files-sdk/openai), and Anthropic's Claude Agent SDK (files-sdk/claude). All share the same file operations and approval-gating defaults, so models can browse, read, and (optionally) mutate your bucket through the same unified surface as your application code. See files-sdk.dev for the current list and per-SDK setup.

MIT