ipfs

ipfs command [options]

ipfs is the command-line interface for the Kubo implementation of the InterPlanetary File System, a peer-to-peer content-addressed storage and distribution network. Each file or block is identified by a CID (Content Identifier) derived from the cryptographic hash of its contents, so the same data is reachable from any node that holds it without requiring a fixed host or URL.A local ipfs daemon acts as a gateway between the user and the global IPFS swarm: it dials peers, fetches blocks, pins content for long-term retention, and exposes both an HTTP gateway and a JSON-RPC API. The add subcommand stores files in the local datastore and prints the resulting CID; cat and get retrieve content; pin add keeps content from being garbage-collected; and files offers a POSIX-like mutable filesystem (MFS) layered on top of immutable IPLD objects.

-r, --recursive

Add directories recursively.

Minimal output.

Wrap files in directory.

init

Initialize local repository.

Start IPFS daemon.

Add file to IPFS.

Download file by CID.

Display file contents.

Pin content locally.

List pinned content.

Show node identity.

List connected peers.

Manage MFS (Mutable File System).

A daemon must be running for most subcommands; ipfs init must be called once per user to create the configuration in ~/.ipfs/. Content is not private by default - anyone who knows a CID can fetch the underlying data through any IPFS node. The default port allocation (4001 TCP/UDP for swarm, 5001 for API, 8080 for gateway) needs to be firewall-friendly for good connectivity.

IPFS was proposed by Juan Benet at Protocol Labs in a 2014 whitepaper. The first reference implementation in Go (later renamed Kubo) reached 0.4.0 in 2016 and 1.0.0 in 2023. A Rust implementation (iroh) and a JavaScript implementation (js-ipfs, later Helia) followed.

curl(1), wget(1), rsync(1)