Welcome to Latest Strikes, your weekly report of the latest Lightning-related news. The last 2 weeks gave us some interesting insights, AI stuff and an LND release. Let's dive in!

River ReportRiver Report

River shared their yearly Bitcoin Adoption Report, with a Lightning section showing Lightning transaction volume skyrocketed in November 2025, whereas number of transactions, while higher than in November 2024, is below November 2023 levels. Sam Wouters from River attributes this to a shift in how Lightning is used, from "gaming and messaging" in 2023 to moving funds in and out of exchanges in 2025. This explains why the transacted volume rose so much, with the average transaction netting $223. Thanks to this, November surpassed a monthly transacted volume of 1 billion dollars.

Methodology: to back this report with data from a network that is private by design, River received aggregated data from some Lightning companies, amounting in total for ~50% of the total Lightning Network capacity in 2025, corrected overlaps, and then extrapolated the data to the whole network. However, Rivers' report doesn't explicitly states how this extrapolation was performed, and simply doubling the figures^[1] would likely result in an overestimation, since the sampled nodes (incl. Acinq and Wallet of Satoshi) represent a very active portion of the network, and it is hence likely that their public capacity drives more transaction volume that the average unit of capacity deployed on the network. Yet, the methodology and polled companies are roughly the same year-to-year, so this is still a very valuable insight when it comes to trend, not absolute numbers.

TL;DR: it's hard to tell whether we really crossed the $1B monthly Lightning volume mark, but Lightning's usage in "economically significant transactions" is definitely going exponential.

StackerNews Discussion

Nostr CoreNostr Core

Pratik Patel and DoktorShift released Nostr Core, a JavaScript/Typescript Nostr Wallet Connect library. It comes as an alternative to Alby's poplar SDK, which includes Alby-related features on top of the bare NWC support. While Alby's SDK is pretty neat for developers looking to easily bring Lightning payments into their app without having to handle custody, I think it's great to have another library available that focuses solely on NWC integration, resulting in a smaller dependency tree.

Another neat part is that Nostr Core more limited dependencies make it available to run on Cloudflare Workers, making it easy (and free) to deploy NWC-enabled apps in the cloud.

GitHub Repository | Medium Post | StackerNews Discussion

L402 DirectoriesL402 Directories

There is a lot of chatter around Lightning being the perfect network for agentic payments (i.e., giving AIs some cash they can use). Of course, Lightning has its drawbacks, but being able to pay an invoice found in a 402 error response and access a paid service is pretty neat. One thing that was missing was a way for agents to discover said services. Currently, a human is usually required to tell the agent what kind of tool to use: grab data from service X ; generate an image through service Y, etc. Satring aims to fix this issue by letting agents search and find L402 enabled services depending on what they're trying to achieve.

It's still pretty early, so there's not much there yet, but the list is being expanded both through active contributions and scrapping. Another missing piece is how agents can tell which service are legit and which are not (i.e. a reputation system), and it actually reminded me of Nostr DVMs (which be covered back in September 2023), where the fact that each service has its own public identity can help foster such a system.

Of course, Lightning x AI isn't new. But we're definitely seeing more and more agents doing payments, and more often than none they're using Bitcoin.

StackerNews Discussion | GitHub Repo

LND v0.20.1-betaLND v0.20.1-beta

Version v0.20.1 of LND was released, bringing several interesting updates:

• LSP detection heuristic in pathfinding: LND nodes will infer from routing hints when an invoice is for a node behind an LSP, and then probe the LSP node directly instead of sending probes to the final destination node.
• several bug fixes, a lot of which address issues at the gossip or network graph level.

Gossip Observer UpdateGossip Observer Update

Jonathan Harvey-Buschel shared an update on his Gossip Observer project, answering some questions and putting forth some early results. He plans to publish a more thorough version on BNOC soon, which we'll cover, but an interesting discovery is the extraction of "communities" (nodes that are tightly connected together) across the network. Two such communities are centered around a seller of preconfigured Lightning nodes, while another is made of a lot of Tor nodes that operate with small channels. An hypothesis that Jonathan is looking forward to evaluate (with one LDK node connected to each community) is whether this communities with few and smaller payment channels lag behind in terms of gossip data, especially since the low capacity of their channels pushes them to connect to a smaller set of "hub" nodes^[2].

1. In Stephan Livera's podcast, Wouters alluded to such a 2x multiplier when extrapolating data from 50% to 100% of the network's public capacity, although for a "number of users" metric that isn't referenced in the report, probably due to how difficult it is to accurately represent such a metric. Quote: "So basically like 2.600000 [users] that we could directly confirm and then sort of extrapolating that out into the network would be publicly around 5.2 million". ↩

2. By hub, I mean nodes with a lot of connections with the whole network. Out of these hubs, few allow nodes to open small channels to them, hence restricting the available peers for small-channels communities. Those peers (such as 1ml) may manage their relatively higher load by not propagating gossip updates as often as others. ↩