A new proposal has been posted to the Firo Crowdfunding System.
- Title: Firo JS SDK Lelantus Integration Toolkit
- Author: emmzyCod
- Proposal type: Wallets
- Link: click here
Use this topic to discuss this proposal!
1 Like
Funding target changed from ‘4.0 FIRO’ to ‘4500.0 FIRO’
Funding target changed from ‘4500.0 FIRO’ to ‘4600.0 FIRO’
Funding target changed from ‘4600.0 FIRO’ to ‘5200.0 FIRO’
While this is interesting, I’m a bit worried this proposal is LLM generated (though happy to be proved wrong). Especially since we don’t use Lelantus anymore and instead Lelantus Spark.
https://www.linkedin.com/in/emmanuel-charles-0b0023250/
Would like to know more details of relevant/similar work to be done.
I appreciate the feedback from the team. In light of being ‘LLM’ generated, it’s not completely false; i carried out the research myself, and the whole proposal was from my own current understanding of the ecosystem, and i used external tools to help with structuring the write-up in a README-style format, and the SDK I’m proposing would be built specifically around Spark’s note structure.
I don’t have direct experience with Lelantus Spark specifically, but the patterns are transferable. From my experience, I’ve worked on developer-tooling projects in other ecosystems like:
- Ox-rollup: L2 dev tools with multi-language SDKs - built Rust core with language bindings for developer tooling
- Arbitrum Stylus VS Code extension: Tooling around a complex cryptographic backend, similar to building interfaces for privacy protocols
1 Like
Hi, I just wanted to make a note on the LLMs and proposals in general before I comment on the proposal. As AI becomes a bigger part of everyone’s writing process—including proposal authors—we can’t expect people to clearly disclose how much they relied on it, nor is there any standard for doing so. So the only practical assumption is that most proposals will involve AI in some meaningful way. What matters is that proposers still own the quality of what they submit, because submitting garbage has always hurt their reputation and will continue to. Reviewers, in turn, should judge proposals purely on their substance, not on how they were produced. With LLMs accelerating both good and bad proposals, I’m sure you’ll see more proposals overall: lots of slop, but also some higher-quality ideas that wouldn’t have surfaced otherwise. The way to handle the volume is for reviewers to use AI tools too—but with humans staying firmly in the loop, using AI to triage and analyse, not to replace judgment.
1 Like
Regarding this proposal, I think it’s a high-impact project. I spoke with CodexEmmzy about shifting the architecture to focus first on a Rust-based SDK with UniFFI bindings that support multiple languages. I’ve worked directly with developers on the Bitcoin Lightning Network and wrote about why we chose this approach a few years ago: https://lightningdevkit.org/blog/ldk-an-sdk-for-the-lightning-network/.
The goal is to make it easy for builders to integrate the FIRO (Lelantus Spark) protocol into their applications without having to reimplement it or understand all the low-level details. High-level APIs can handle the heavy lifting. In the age of AI engineering, if FIRO can provide usable yet secure tools and SDKs, it could spark thousands of new experiments and applications. With AI engineering plus the FIRO SDK, someone should be able to build a Campfire-style wallet in a weekend.
This is a large and important project that deserves the core team’s attention, so I’d love to hear their feedback. I’m also able to provide advisory support and can personally contribute up to 50% of the funding as a donation.
2 Likes
The idea is great for me, but I have some questions in my mind that I was not able to find answers for when reading the proposal. First of all:
- Please provide a technical design description, project components, the meaning of each component, and the functionality of each.
- Please give a description of the sdk’s functionality. For example: Will it provide wallet functional or just tx creation?
Thanks for the feedback! As for the technical design, I’ve edited it to include that part. I also have a technical architecture document (kept separate because I did not want to burden the main proposal) which gives a lot more information on how the technical flow works.
Full Technical Architecture: Technical Architecture Document
In the proposal you can see what the SDK does and what’s reserved for wallets. The architecture doc goes deeper into component communication, data flows, and integration patterns.
We’ve reviewed the document you provided and again looks AI generated. While we aren’t against AI assistance, the doc also mentions that you want to implement Lelantus Spark proofs in Rust which is a huge undertaking in itself and our devs don’t think this shows sufficient understanding to make this all work.
A significant portion of the TypeScript SDK has already been written and tested. Spark is next on the list while I currently work on a separate branch. Although my pull request was not merged, I can respect preserving BitPay’s initial work started using JavaScript 12 years ago. My implementation is and will be completed as a full TypeScript SDK, which removes that need and lays significant ground work for any future SDK using any other language. Unfortunately, I never had time to learn Rust, but I could do this in Java if that was of interest. I will continue to publish the library using my account on NPM, but wouldn’t be opposed to providing the full implementation to the Firo team if a proposal was sponsored, or you can just fork it as you see fit.
1 Like