Messari published their State of Stellar Q1 2026 report ahead of this meeting, and there are numbers in it that every developer building on Stellar should understand — not as vanity metrics, but as signal for where the ecosystem is actually heading and where the real opportunities are.
RWA market cap — $2 billion at the time of the meeting, up from approximately $1.5 billion at the end of Q1 2026. That is roughly a 33% increase quarter over quarter and a 3x increase year over year from the $500 million mark in Q1 2025. The biggest single-quarter move across any metric in the report.
Average daily smart contract volume — $16 million per day in the last quarter. One year ago it was $2 million per day. For five or six consecutive quarters, Soroban transaction volume has only gone up. This is real usage: contracts being called, operations executed, applications that real users are touching.
Stablecoin market cap — $300 million, up 20% quarter over quarter. Protocols driving this include Aquarius, Blend, and SolarSwap.
The synthesis: Stellar is executing a compounding flywheel that is now clearly visible. Institutional capital flows in via RWAs, higher TVL attracts more DeFi protocols, more DeFi protocols drive more Soroban invocations, more on-chain activity attracts more developers, more developers build more user-facing applications, more users generate more stablecoin and payment volume — and all of that makes Stellar more compelling for the next wave of institutional adoption.
The full report is on messari.io and is worth reading in detail.
Announced on April 22nd at Stellar House in Mexico City: MoneyGram and the Stellar Development Foundation have announced a multi-year extension of their partnership. The framing is important — this is about scaling real-world stablecoin utility globally.
This partnership has been running since 2021. In that time they built the world's largest cash on- and off-ramp for digital assets, launched the MoneyGram Ramps API so developers can plug into the network directly, and created a stablecoin balance feature inside the MoneyGram app that lets customers hold funds in digital dollars and cash out at physical locations.
The next phase focuses on Latin America. The MoneyGram app with stablecoin support first went live in Colombia. With this announcement, it is extending to El Salvador, with more countries across Central and South America planned throughout 2026. The stack powering this is Stellar, Cross-Mint, and Circle's USDC.
The real-world scale here: MoneyGram operates nearly half a million retail locations across 200+ countries. That is not a sandbox — that is live infrastructure for people sending money to families and communities that depend on cash-based services. Stellar Development Foundation CEO Denelle Dixon said: "Stellar was built on the belief that the global financial system should work for everyone." This partnership is what that looks like in practice.
Aquarius, the decentralized liquidity management platform built on Stellar, crossed $50 million in total value locked. Aquarius is Stellar's answer to DeFi liquidity infrastructure — AMMs, liquidity pools, reward systems — and it is designed to be a foundational liquidity layer that other Stellar protocols build on top of. When Aquarius TVL grows, the entire Stellar DeFi ecosystem becomes more liquid and more efficient.
For developers asking where to start: Aquarius has been investing in documentation. Their docs at docs.aqua.network now include real code examples — how to execute swaps through the optimal path, how to claim LP rewards, how to add fee configurations to swaps for building routers into existing applications. If you are building anything on Stellar that involves swapping assets, yield strategies, or liquidity-adjacent features, the Aquarius docs are a solid reference for what production-quality Stellar DeFi integration looks like.
The Stellar Agents hackathon on DoraHacks just wrapped, with around 600 hackers submitting over 260 projects. The hackathon was focused on agentic payments using x402 and MPP — the two protocols covered in the previous developer meeting. What the hackathon showed is what happens when you put those primitives in the hands of 600 builders from around the world.
The following are standout submissions, highlighted not to endorse any specific project but to use them as a window into what is actually possible to build on Stellar right now.
stellar8004.com addresses one of the most important unsolved problems in the emerging AI agent economy: there is no standard way to find agents, verify who built them, or evaluate whether they are reliable. You cannot search for a trustworthy A2A agent for a given task and get a verifiable answer. Every platform has its own walled garden, and there is no portable reputation.
Stellar 8004 takes its name from ERC-8004 on Ethereum (now going multi-chain) and gives every AI agent a blockchain-backed identity, a discoverable endpoint, and a reputation system you can verify on-chain. Think of it as LinkedIn for agents, except the reputation is not self-reported — it is attested on-chain.
The stack is built on three Soroban contracts:
The live explorer at stellar8004.com lets you browse registered agents, check reputation scores, interact with them, and leave feedback. Registered agents can also advertise x402 and MPP payment support in their metadata, so agents can be discovered and paid per query in USDC.
For builders, the team also published a TypeScript SDK at tryon.lab/stellar8004 and a Claude Code skill with slash commands for building on top of the registry with AI assistance.
The conceptually powerful part is the agent-to-agent trust loop. When a DeFi agent wants to call a data analysis agent, it currently has no way to verify that agent is legitimate or reliable. With a registry like this, it can query on-chain, check the score, verify attestations, and decide whether to proceed programmatically.
One note for anyone building in this space: distribution is everything for infrastructure like this. In just one hackathon there were four or five similar projects targeting the same use case. If you are building an agent registry or identity system, check the ecosystem page first and understand the competitive landscape.
mpprouter.net.dev is a payment network router specifically designed for agent-to-agent micropayments. On Tempo, there are over 400 MPP endpoints. This router makes it possible for users and agents on Stellar to reach and use those endpoints by paying on Stellar.
Not yet on mainnet, but the problem it solves is real and concrete: 400 agents on another network, and this project makes them reachable from Stellar. The way to find good project ideas is to find the pain points of a network — this one found a clear one.
With over $3 billion lost to hacks and exploits in 2025 (the worst year on record), and the industry already down over $600 million in just the first weeks of 2026, autonomous smart contract auditing is becoming genuinely important infrastructure.
Built under the handle Chinese Powered, this is an autonomous AI security auditor for Soroban smart contracts. It works in three steps:
The insight that makes this composable is step three: by writing audit results on-chain, you create an auditing layer that other contracts or agents can query. The project is currently on testnet and needs more work before it is production-ready — particularly around trust establishment and distribution — but the vision it represents is one of the most important unsolved problems in DeFi.
Over 75% of crypto users access services via mobile, and in the markets where Stellar is most relevant — Latin America, Africa, Southeast Asia — mobile-first is not a preference, it is the only option.
Built by Sam Felix, this project flips the assumption that running an AI agent requires cloud infrastructure. Take an old Android phone, run a FastAPI agent runtime on it using Ollama for local inference, and connect it to Stellar via MPP for per-request micropayments. Anyone with a phone can become a node in the AI agent economy.
What is particularly clever is the no-code layer on top: a visual drag-and-drop builder where non-technical users can design agent workflows without writing any code. The exact use case is letting someone in an emerging market use their device to provide or access AI capabilities that would otherwise be inaccessible or expensive. Not yet on mainnet, but the direction is worth noting for anyone building for global audiences.
Highlighted as one of the most polished projects from the CDMX hackathon, Talos is an autonomous agent corporation framework. It lets you create sub-agents and orchestrate them to act as corporate employees: dedicated agents for GTM strategy, marketing campaigns, engineering lead hiring, and so on. The payment layer runs on x402. If you have worked with Paperclip or similar autonomous agent setups before, the mental model will be familiar. One of the most common questions at the CDMX hackathon was how to use AI to steer an entire project or company — Talos is one of the more complete answers that came out of that event.
Stellar House is SDF's format for bringing together builders, partners, and ecosystem members in a physical space — typically two to three days of talks, workshops, and demos. The Mexico City event just wrapped, and all sessions were live-streamed on the SDF YouTube channel (find them under the Playlists section of the channel).
Mexico City was particularly significant for two reasons: the MoneyGram announcement came out of it, and Mexico and Latin America broadly are central to Stellar's mission. The event also featured pitch presentations from several teams that came through the CDMX hackathon — building a pipeline from hackathon to Stellar House is a format that seems to be working.
For developers building on Stellar with AI tooling, a new open-source resource is available at github.com/kaankacar/stellar-ai-guide-mx. Check the building with AI section at developers.stellar.org for the full set of AI resources, including the llms.txt file for feeding Stellar context into LLMs and Stella, the AI assistant trained on Stellar documentation.
Protocol 26, codenamed Yardstick, went live on testnet on April 16, 2026 at 17:00 UTC — literally the day of this meeting. Mainnet validator votes are scheduled for May 6th. If you are running any kind of infrastructure on Stellar — nodes, SDKs, integrations — now is the time to pull the Protocol 26 changes, update your SDKs, and test against testnet before the mainnet vote. The #protocol-next channel in the Stellar developer Discord is where the ecosystem coordinates during upgrades.
Two CAPs shipped with this release:
CAP-81 rewrites the Soroban eviction scan. Instead of scanning the BucketList on disk, eviction now works directly from in-memory Soroban state. The result is faster evictions, fewer disk reads, and more efficient node operations. This is the kind of under-the-hood improvement that compounds as Soroban usage grows — less I/O overhead means the network handles growing state more gracefully at scale.
CAP-82 adds checked arithmetic variants for the existing 256-bit integer host functions. Where the current functions trap on overflow, the new checked variants return Void instead, letting contracts handle arithmetic errors explicitly and gracefully rather than panicking. If you're writing DeFi contracts — lending protocols, swap routers, pricing curves — this matters: checked arithmetic means operations fail explicitly and predictably instead of silently wrapping or trapping.
Stellar has crossed $2 billion in tokenized real-world assets on the network — more than 4x growth in 12 months and 2.5x growth year-to-date as of April. This is not speculative TVL. These are actual financial instruments: tokenized treasuries, bonds, and private credit sitting on-chain today. The institutional names driving this are serious: Franklin Templeton's tokenized treasury fund, Ondo Finance, RedSwan Digital, and Centrifuge. These are not crypto-native experiments — these are traditional finance institutions choosing Stellar as their settlement layer.
What makes this meaningful for developers is that those $2 billion in assets do not just sit there. They are becoming collateral in lending protocols, being deployed in DeFi, and earning yields. That creates real primitives to build on top of.
The Stellar DeFi ecosystem is maturing quickly. Here is the current state:
Blend has crossed $80 million in TVL, making it one of the most significant lending protocol deployments on Stellar.
Stellar Templars Protocol added six new real-world assets to lending markets on Stellar. You can now borrow stablecoins against tokenized institutional assets. The newly added collateral types include:
Institutional-grade RWAs as DeFi collateral is a significant milestone. Sovereign debt meeting DeFi accessibility is a phrase worth sitting with.
Sushi deployed their first concentrated liquidity DEX on Stellar. PYUSD/USDC and XLM/USDC pools are live. Cross-chain functionality is coming soon through Squid Router, which connects Stellar assets to 80+ other blockchains.
EURA — a MiCA-compliant euro stablecoin from All Unity — is live on Stellar. This is a fully backed, regulated euro stablecoin built for institutional use. A regulated euro stablecoin on Stellar is a notable addition for builders targeting European markets or building compliant financial applications.
For a full picture of where Stellar DeFi stands, the SDF published a post titled "What the DeFi is happening on Stellar?" that captures how the ecosystem has been quietly maturing while others debate narratives.
Both x402 and MPP — the two leading agentic payment protocols — are now live on Stellar mainnet. This is a significant milestone for anyone building at the intersection of AI and payments.
The entire payment infrastructure of the internet was designed for humans. An AI agent that needs to call a paid API hits a fundamental wall: there's a form, a billing dashboard, an API key a human configured. The agent hits a paywall and the workflow breaks. x402 and MPP solve this by making machine-to-machine payments as simple as an HTTP request.
x402 is built on the HTTP 402 "Payment Required" status code — a code that has existed since the 1990s but was never meaningfully implemented until now. The flow is simple: an agent requests a resource, the server responds with a price, the agent authorizes stablecoin payment, and the resource is delivered. One HTTP round trip.
On Stellar, this settles in under 5 seconds. Stellar's fees — approximately 0.00001 XLM per transaction — are essential here: if your transaction fee costs more than the payment itself, micropayments do not work. Stellar also has USDC, PYUSD, and USDY as first-class native assets — not bridged, not wrapped.
x402 is co-governed by Coinbase (who launched it), Cloudflare, Google, and Visa. Google has integrated it into their agentic payments protocol. Cloudflare built it into their pay-per-crawl tooling. The SDF also brought OpenZeppelin into the picture — they secure over $130 billion in on-chain assets — to power the x402 facilitator and provide audited smart account contracts with programmable spending limits and guardrails.
MPP was co-developed by Stripe and Tempo and launched last month, and it is live on Stellar. Where x402 handles one payment per request, MPP introduces sessions: an agent pre-authorizes a spending limit up front, then streams micro-payments within that session, settling in bulk at the end. If your agent is querying a data feed thousands of times an hour, MPP sessions are the right model — you avoid the overhead of a per-request authorization for every single call.
On Stellar, MPP settles via Soroban SAC transfers. Agents never need to hold gas tokens because the SDK handles server-sponsored fees.
MPP launched with over 100 integrated services including Stripe, Anthropic, OpenAI, Shopify, and Visa. The core spec has been submitted to the IETF for standardization as the official HTTP payment standard.
To connect this back to everything else in this meeting: those DeFi protocols, tokenized assets, and stablecoin rails all become accessible to AI agents through these payment primitives. An agent can borrow against a tokenized treasury, pay for a data feed, execute a swap, and settle cross-chain — all programmatically, without a human in the loop.
If you want to start building with these protocols, the best starting points are:
llms.txt file for feeding Stellar context directly into LLM prompts, and Stella, the AI assistant trained on Stellar docsThe second half of this meeting featured Yahya and Mert, co-founders of Noether from the No Ethers team, recent graduates of the latest Stellar Community Fund funding round. Noether is, to their knowledge, the first perpetual DEX on Stellar.
The opportunity is straightforward: every serious DeFi ecosystem needs a perpetual DEX, and Stellar did not have one. Mert has been building on Stellar since smart contracts launched on the network, has hosted over 10 Stellar workshops in Turkey, and has mentored builders and served as a judge in Stellar hackathons. When Yahya identified the gap, they locked in and built.
Mert walked through a live testnet demo of Noether:
Faucet — A testnet faucet page where users can claim up to 1,000 USDC per day. Before claiming, users need to set up a trustline to the testnet USDC asset — a standard Stellar requirement for holding any non-native asset.
Trade Page — The core trading interface has three components:
Smart Contract Status — The contracts are currently at MVP stage. Mainnet launch is planned approximately two to three months out, following a smart contract audit that will be supported by the Stellar Development Foundation. The team emphasized they will not launch on mainnet before the audit is complete.
When asked what percentage of their codebase was written with AI assistance, Mert estimated around 70% — with an important caveat: the core financial logic in smart contracts was written and verified by hand. Their approach was to use AI for the less critical parts (landing pages, boilerplate, tooling) while personally reviewing and proving the correctness of anything financially sensitive. As Mert put it: "We write the code with AI but we read it."
]]>In this protocol meeting we had two CAPs to discuss and that was CAP-81 and CAP-82.
CAP-81: This CAP introduces a simpler and more efficient eviction scan process. Instead of scanning the BucketList, the new eviction scan just uses in-memory Soroban state. This simplifies the logic while reducing disk reads and speeding up eviction speed.
CAP-82: This CAP adds checked variants of the existing 256-bit integer arithmetic host functions. Unlike the existing functions which trap on overflow, the checked variants return Void on overflow, allowing contracts to handle arithmetic errors gracefully.
Read more about the proposals here:
]]>We have one new CAP to discuss and that is CAP-80. This proposal adds BN254 Multi-Scalar Multiplication and modular arithmetic used in a variety of ZK proof applications. Adding host support for these will greatly improve the performance of these use cases.
We also follow up on a previous CAP - the CAP-73.
Read more about the proposal here:
]]>In this year’s first Stellar Developer Meeting, on Thursday, January 22nd at 8am PT on Twitch, we will catch up with Ozgun and Boyan from the OpenZeppelin team and take a look at the libraries released in Q4 last year.
We had three CAPs prepared for this meeting. CAP-77 will provide a way to make ledger keys inaccessible based on the network configuration upgrade performed with a validator vote. CAP-78 proposes an interface which allows developers to specify TTL extension policies such as 'if an entry TTL is less than 29 days, extend it to have 30 days TTL'. CAP-79 introduces host functions for converting Stellar strkey format strings to/from Address/MuxedAddress objects.
Read more about the proposals here:
]]>In this meeting we do a walkthrough of OpenZeppelin’s Q3 library releases for Smart Account, Vault and RWA.
Links to docs:
]]>In this meeting we are continuing our mini series about the Stellar-based open source tooling OpenZeppelin is developing. In the last session we talked briefly about Relayer and this time we are diving deeper into this tool, and OpenZeppelin Managed Service.
OpenZeppelin Relayer: https://docs.openzeppelin.com/relayer
]]>In this meeting we are talking to Dobprotocol to learn more about how they use Stellar to build a unique platform, turning machines into investable assets.
Website: https://www.dobprotocol.com
]]>In this call the recent state archival issue, introduced by Whisk (Protocol 23), is discussed. Questions from community builders are also answered.
]]>We are meeting with the OpenZeppelin team again to catch up on the latest developer tooling work they are doing.
This time Steve will demo UI Builder, an easy way to spin up a front-end for any contract call in seconds. UI Builder allows you to select a function and then it auto-generates a React UI with wallet connect and multi-network support, and exports a complete app.
In this Core Advancement Proposal discussion CAP-0075 (Host functions for enabling Poseidon and Poseidon2 hash functions) is presented.
This CAP proposes adding host functions for cryptographic primitives enabling Poseidon family of hash functions, which are widely-adopted hash choices in efficient zero-knowledge proof systems. Supporting these as host functions in Soroban can facilitate adoption of ZK applications and interoperability with other ecosystems.
Link to CAP-0075
]]>In this meeting we will discuss CAP-74, BN254 and Poseidon hash functions.
]]>In this week’s Stellar Developer Meeting we discuss new Core Advancement Proposals.
Here are some resources to read up on:
]]>In this meeting we talk to Flashback founder Brieuc Berruet, who participated in the DraperU x Stellar incubator program last year, to get an update on his project (spoiler alert - it's live!).
Learn more about the project, and see a cool demo!
]]>In this meeting we are talking to Esteban and Francisco from PaltaLabs. Stellar Hacks is currently running a hackathon in collaboration with PaltaLabs, where the goal is to build using their projects Soroswap and DeFindex.
Let's see how the hackathon is going, and some advice if you are still considering entering, there's still time to enter!
Links:
]]>In this meeting the Nomyx team will give us an introduction to their advanced smart contract architecture (an implementation of the Diamond Pattern) and the Nomyx Diamond proxy viewer interface, that facilitates inspection of deployed diamonds and makes them a bit easier to work with.
They will also be showcasing a live deployment instance of the diamond proxy standard currently used for the Nomyx platform which removes limitations related to the upper contract size limit, allowing developers to launch complex products.
]]>In this call we are getting an update on the exciting Scaffold Stellar project by Fifo from Aha Labs.
Scaffold Stellar is a great way to fast-track new projects built on Stellar. It has starter smart contracts, frontend, plugins and tooling built-in.
]]>At this Stellar Developer Meeting we are having a chat with Bram Hoogenkamp from OpenZeppelin about the OpenZeppelin Monitor - providing monitoring and alerting for smart contracts.
]]>We are close to completing the implementation of Whisk, Protocol 23 CAPs, the next step in the CAP lifecycle is voting by the Core CAP team. In this protocol meeting an update was provided on the CAPs and any final questions from the community was answered.
Read up on the CAPs here:
]]>This week we’ll dive into one of the most overlooked tools - Quickstart.
Quickstart is a local Stellar network environment (node) that allows developers to run a local version of the Stellar network for development and testing.
Learn about Quickstart, see how to install it and how to deploy the Hello World sample dapp on Quickstart.
Link:
]]>Pamphile (better know as tupui) is a senior software engineer working at Bitpanda. Previously he worked in the team who brought Flight Simulator 2020, and he created a consulting company to work on open source software.
Now he is building his project named Tansu on the Stellar blockchain. Tansu is a decentralized system which aims to secure the software supply chain and break siloed communities by using a DAO.
Link:
]]>