DCF

Big milestone for the Clarity Act yesterday. While everybody is talking about what it's going to do for stablecoins and economies, the bigger win (IOHO) is the potential for blockchain as a whole. Good read here from @a16zcrypto that speaks to the wider implications of the eventual full passing of the act. "What builders need to know about the CLARITY Act, what it is and why it matters" a16zcrypto.com/posts/article/…

Where does HardenedJS sit in the JavaScript stack? Three primitives. Lockdown freezes the intrinsics. Harden freezes the transitively reachable object graph. Compartment runs untrusted code with only the authority you grant. That is the surface.

If it seems that @Ymaxapp is taking the hard work out of manually searching chains for the best yield, that's because it is. One signature - multiple chains. @agoric Zoe contracts and Orchestration are at the heart of how it works. You'll be hard-pressed to find infrastructure anywhere else that can do this as seamlessly and as safely as Agoric.

Most AI agent frameworks work like this: put all your credentials, all your keys, all your authority in a bag, and hand the bag to the agent. That framing should stop you cold. If it can read your calendar, it can probably read your email. @__Endojs__ grants only what the task needs, nothing more.

Another week, another ambient authority hack. The carrier keeps changing (Morse, base64, HTML comments, hidden Unicode) but the shape hasn't agents holding broad authority gated by policy a refactor can delete. This is what capability discipline is for. This is safety that @__Endojs__ provides.

🫵Hey, you! Why do you DeFi? For us it's simple. Capital should work as hard as the person who earned it, and you shouldn't need a computer science degree to make it happen, regardless of what chain it's on. What about you?

Most chains hope their contracts are safe. @Agoric uses @__Endojs__ to make sure of it. Object capabilities, hardened compartments, no ambient authority. Smart contracts that can't reach what they weren't given. This is technology 35 years in the making. Thanks to the need to keep AI agents in their place, it's about to have its moment.

Our bonus recommendation this week is this useful contribution to understanding how blockchain becomes institutionally meaningful. “Blockchain for Good” shows how blockchain ventures mobilize imagined possibilities for action to build a field around social impact. Author: Paula Ungureanu 🔄 Read it here: sciencedirect.com/science/articl…

A new governance discussion has started @agoric Discourse. Topic: Increase the deposit amount required to execute an on-chain governance proposal. Head over to Discourse and be part of the conversation. community.agoric.com/t/re-secure-sy…

Another day, another ambient authority disaster. All avoidable with @__Endojs__ and object-capabilities.

That sucks! This is what ambient authority looks like at scale. This is the sort of thing our team @__Endojs__ has built to mitigate. The agent found a root token and used it........System prompts can't stop that........but architecture can. With object capabilities, the agent only gets a reference to what it needs. The Railway API doesn't exist for it. You can't delete what you were never handed. github.com/endojs/endo

One Signature Portfolios are live on Ymax. Land on the app. Pick a path. Sign once. You're deployed across multiple protocols and chains 🤯 That's the whole flow. Pick what suits your approach - Curated Strategies: deposits routed through third-party managed vaults across multiple protocols. More moving parts, higher yield potential. Direct Lending: deposits placed straight into established lending pools. Simpler, more familiar. The magic is what happens after you pick. No manual bridging. No chain switching. No confirming five separate transactions. One signature kicks off the whole thing. Ymax handles the rest on-chain. Try for yourself: t2m.io/aD9igQE

We've just published an important blog article. 📍Containing AI Agents: The Endo Familiar in Action📍 Three questions for anyone shipping AI agents: 1️⃣ If the model gets prompt-injected, what can it reach? 2️⃣ If it tries to exfiltrate data, what stops it? 3️⃣ If a dependency is compromised, what's the blast radius? 🫨 If those answers make you nervous, they should. The good news: there's a foundation built for exactly this problem. Watch @agoric 's @kriskowal demo the ⭐ Endo Familiar ⭐ , plus our latest post on why containment has to come first: lnkd.in/exyVNftH

The @axios hack highlights once again the important work being done by @__Endojs__ Check out their latest blog post covering the attack and ways that it could be prevented in the future.

A new @agoric proposal is live Agoric - Proposal #118 Text Proposal: Introduce a Deflationary Mechanism – Burn 1 BLD per Reward Claim 📋 TL;DR: Summary This proposal introduces a simple and transparent deflationary mechanism for the Agoric ecosystem: Burn 1 BLD every time staking rewards are claimed. The goal is to: Introduce a predictable and consistent supply sink Align network activity with the value of BLD Strengthen the long-term sustainability of the token economy Voting Ends: April 2, 2026 at 8:07 PM UTC

Human cooperation is happening here. Mark your calendars for June 14. @web3privacy

According to @mapofzones @Agoric is currently #1 in 7-day IBC volume across the entire Cosmos ecosystem!! $571K moving cross-chain. That puts it ahead of Kava, Terra, UX Chain, and every other zone on the map. What's more interesting is the direction. Of nearly 2,000 IBC transfers this week, 98.8% were inbound. Assets are arriving, not leaving. For a chain that doesn't make much noise, the on-chain activity tells its own story and @Ymaxapp is the star character.

@agoric's @Ymaxapp solving the execution challenge for optimizing yield opportunities on stables. ONE Signature. Just one. Replacing what would have historically taken up to 12. That's less friction to better yield.

@_weidai The EndoJS project is building a p2p, multiplayer agent harness that manages permissions with a hierarchical petname system, such that the AI is not exposed to even its own credentials, just self-assigned names. DM - would love to chat.

Who's working on "secure" agent harnesses? Ingredients: track all tool calls, semantically label them (what was read, what is the effect), enforce policies, etc. Would love to chat.