Something rare is happening for NEAR right now and I think we have weeks, not months, to act on it.
In the past week, Beanie (blknoiz06), Trustless State, and Arthur Hayes have all spoken about NEAR to their combined millions of followers. NEAR Intents has crossed $10B in cumulative volume, with around $2.1B in the last 30 days alone. NEAR AI shipped Ironclaw — the security-first, WASM-sandboxed, TEE-deployed alternative to OpenClaw. Each of these matters on its own. Together they’re the closest thing NEAR has had to a coherent narrative moment in years.
But narrative without builders decays. Mindshare is borrowed; product is earned. If the next 30–60 days don’t produce visible Ironclaw agents that ordinary crypto users can point at and say “this is what NEAR does now,” the attention moves on.
So the idea I want to put in front of this community is simple: money attracts talented builders, and right now is the moment to use it. A fund — call it the Ironclaw Builder Fund — to bring capable developers into NEAR specifically to ship Ironclaw agents that use NEAR Intents.
Why Ironclaw specifically
Ironclaw isn’t just another AI thing. The exact reason OpenClaw users have been losing credentials is that current-generation AI agents demand a level of trust they shouldn’t get. Ironclaw’s design — credentials in an encrypted vault, WASM-sandboxed tools, TEE deployment — removes the trust requirement instead of asking users to extend more of it.
That positioning maps perfectly onto crypto users. The people who’d never hand keys to an OpenAI-hosted agent are exactly the people who’d use an Ironclaw agent that executes trades via NEAR Intents while keeping their wallet private. The PMF is more obvious here than in mainstream AI.
But obvious PMF doesn’t ship products. The pool of independent builders working on privacy-focused AI agents is small, and most aren’t crypto-native yet. A fund is the forcing function that pulls them in while the narrative is hot.
The marketplace this builds toward
The longer-term picture isn’t just “twenty Ironclaw agents shipped.” It’s a marketplace. Builders publish agents, users discover and install them, payment flows through NEAR Intents, and TEE deployment means trust lives at the platform layer rather than each agent having to earn it from scratch.
If that marketplace exists, the flywheel becomes obvious: more agents bring more users, more users attract more builders, NEAR Intents captures economic value from every agent transaction, and NEAR becomes the natural home for privacy-first AI agents the way the App Store became the natural home for mobile software. The fund alone doesn’t build the marketplace — but it creates the agent supply that makes building it worthwhile.
Leading from the front
The strongest signal NEAR could send isn’t a fund announcement on its own. It’s a fund announcement alongside great Ironclaw agents already shipped by the people closest to the project. Illia, George Zeng, the NEAR AI team, NEAR Foundation leadership — agents from them set the bar for what “great” looks like and signal to external builders that this is the ecosystem’s direction, not a side experiment.
A fund attracts outside builders. Internal builders shipping their own agents is what makes those outside builders take the fund seriously.
What I want to start discussing
I want to keep this thread open and exploratory before getting into mechanics. A few questions I’d love views on:
-
What kinds of Ironclaw agents would actually move the needle for NEAR?
-
What does “successful” look like — agents shipped, Intents volume driven, builders retained in the ecosystem?
-
What needs to be true today for an Ironclaw agent marketplace to emerge naturally tomorrow?
Funding mechanics are downstream. The bigger question — and where I’d like this thread to focus — is how we as a community should approach the process of creating great AI agents on NEAR. That direction sets everything else: what to fund, who to involve, how to measure success.