Smokin Ted
18.5K posts

Smokin Ted
@_SmokinTed
Research & Strategy @TailoredWeb3







Despite a challenging start to the year for DeFi and yield-bearing assets, @maplefinance continues to stand out. Both syrupUSDC and syrupUSDT have continued to grow, with their respective supplies increasing by over 33% and 36% year-to-date. Their supplies are standing at around 1.55B for syrupUSDC and 0.9B for syrupUSDT, placing both assets among the top 10 largest yielding assets in the market. Deposits into these products have also continued to rise, reaching an ATH of $2.5B and still growing. This sustained momentum is not surprising: for over a year, syrupUSDC and syrupUSDT have delivered some of the highest and most consistent yields compared to their competitors. Below is a breakdown of supply across the different chains where these assets are deployed. As illustrated, their cross-chain expansion has been a key driver of growth. Additional deployments are expected in the near future. Maple.





An AI agent can now on the internet make payments even with no API key. This is now possible through @stripe's new Machine Payments Protocol (MPP) which is an open standard for how AI agents and machines pay each other online. MPP is an open standard rather than a proprietary Stripe-only product. — Think of MPP as “payments for the agentic internet.” Instead of subscriptions or API keys, services can be paid per-use, per-second, or per-task automatically. MPP is an application-layer protocol for programmable payments between machines. It is a protocol that lets one software actor request payment from another, with enough structure to support different underlying rails. In plain English, MPP is trying to solve how an agent can discover a service, learn the price, authorize payment, complete the payment over whatever rail is available, and receive the resource or result all without manual signup, checkout screens, API keys, or a human in the loop? This means that there is no account setup and no API keys . Rather it boasts of: • per-use access • support for agentic commerce • use cases like data queries, • communication APIs, search, research, A/B testing, feedback, and human task marketplaces Notably, MPP is payment-rail agnostic with stablecoins, cards (via @stripe/ @Visa ) and even Bitcoin supported from day one. — MPP’s strongest utility is in situations where normal SaaS billing breaks down with early use cases already live: ● Per-call and per-query services Instead of registering an account and adding a card while storing an API key or getting a monthly invoice, an agent can: • discover a service • pay on demand • use it once and move on This is ideal for search, scraping, data, inference, analytics, routing, and orchestration. ● Cross-provider agent toolchains An autonomous agent may use ten external services in one job. MPP would let it: • pay each provider directly • choose based on price/performance • switch dynamically without human account provisioning • keep auditable machine-readable payment trails That is much closer to how cloud compute is consumed programmatically. ● Micropayments and streaming Many machine interactions are too small or too dynamic for subscriptions: • fractions of a cent per request • per-token model usage • per-second communication • per-byte delivery • dynamic metered access MPP seems designed for these cases. ● Human-in-the-loop task markets With MPP, agents can hire humans for feedback, testing, social tasks, and A/B tasks. That extends MPP beyond APIs into a broader labor marketplace. In that model: • the buyer is a machine • the worker may be human • the payment commitment still needs machine-readable enforcement That could be a large category if agents increasingly outsource specialized edge cases. @tempo’s posts also mention future enterprise features and use cases beyond agents: • global payouts • remittances • embedded finance • tokenized deposits All these suggest that MPP may become a general protocol for programmable treasury and operational payments, not only AI agents. — Here’s a simplified version of MPP's architecture : • Service advertises price • Agent creates payment intent • MPP routes to chosen payment method • Payment proof is verified • Access is unlocked (API, data, service, etc.) — Stripe’s involvement matters because Stripe brings the strongest bridge from internet-native developer workflows to mainstream payments infrastructure. Stripe seems to be betting that with time the real market will be broader: • some machine payments will use stablecoins • some will use cards • some will use Bitcoin flows • enterprises will want compliance, abstraction, retries, reporting, and familiar payment primitives. Knowing this, developers will want one protocol, not a separate integration for each rail. That is the strategic significance of MPP. The biggest problem in agent commerce is economic interoperability and MPP is one solution to this problem. h/t : @jeff_weinstein ( Introducing MPP ) @dwr ( Rails and extensibility ) @uttam_singhk ( Comparison with x402 ) @cuysheffield @matthuang


One well-crafted prompt beats hours of manual DeFi. This video teaches you exactly how to prompt on INFINIT. Watch this and you're already ahead in the Prompt-to-DeFi Strategist Challenge. 👇














$1.7B+ for syrupUSDC and $860M+ for syrupUSDT. First, syrupUSDC and syrupUSDT crossed $2.5B in combined deposits. Now, both dollar yield assets have set new individual deposit ATHs.



1/ Looping makes up ~1/3 of all Ethereum lending activity. Most of it is staked ETH and stablecoins. RWA looping is next, and it changes everything. It’s the biggest DeFi strategy most people still aren’t talking about. Here’s how it works and how ynRWAx fits in 🧵
