Radicle is an open source, peer-to-peer code collaboration stack built on Git. Unlike centralized code hosting platforms, there is no single entity controlling the network. Repositories are replicated across peers in a decentralized manner, and users are in full control of their data and workflow.

I know that a number of projects are working on github replacements on nostr, but this seems to be a different way of creating a more p2p github replacement.

I didn't look at their code, and the FAQ was not terribly enlightening to me.

The protocol page has more detail:

Radicle adopts a local-first, peer-to-peer (P2P) architecture, which draws inspiration from Secure Scuttlebutt (SSB) and Bitcoin’s Lightning Network.

Nodes on the Radicle network subscribe to repository data they are interested in, and peers announce updates that in turn trigger fetches for the underlying content. Just like SSB and Lightning, updates are gossiped on the network until they reach all interested peers.

Peer connections in Radicle are secured thanks to a Noise protocol handshake. Radicle uses the Noise XK pattern specifically, just like the Lightning Network with the Node ID as the static key. This requires nodes to know the Node IDs of their peers before connecting to them, which takes place through the exchange of peer information over the gossip protocol.

Unlike SSB’s focus on social networking via append-only logs, Radicle focuses on code collaboration by incorporating Git’s object model and transfer protocol into a peer-to-peer context. This architecture not only leverages Git’s proven efficiency and reliability but also gives users complete autonomy over their social artifacts. Radicle’s peer-to-peer architecture, in contrast to federated systems, ensures no centralized points of failure, allowing the network to persist as long as users operate nodes.

Here's a more general description of the protocol:

Heartwood, the latest generation of the Radicle protocol establishes a sovereign data network for code collaboration and publishing, built on top of Git. In Radicle, users maintain local copies of their repositories of interest and related social artifacts such as issues and patches. Instead of depending on a centralized service like GitHub, each participant in Radicle operates a node that is capable of running on a personal computer, and is connected via a peer-to-peer network.

Nodes, identified by public keys, host and synchronize Git repositories across the network, using a novel gossip protocol for peer and repository discovery, alongside Git’s protocol for data replication. In combination, this allows nodes to locate, replicate, and verify any repository published to the network, provided at least one other peer seeding the repository is online. Since Radicle is built on Git, it can easily interoperate with existing tools and workflows.

Radicle’s architecture is local-first, ensuring continuous access to one’s repositories directly from their device, regardless of internet connectivity. Repositories have unique identifiers and are self-certifying, meaning all actions, from committing code to adding a comment to an issue, are performed locally and cryptographically signed, allowing peers to verify authenticity and data provenance once propagated to the network. This allows trust to be established without reliance on a centralized authority.

Radicle is designed for extensibility, allowing for diverse use cases without necessitating modifications at the protocol level. This guide delves into the capabilities of Radicle’s initial release, which has a focus on code collaboration and code publishing. Nonetheless, a range of other applications is foreseen in the future and possible today, including knowledge sharing, project coordination, and data set collaboration.