mcpmessenger

Quick update for the technical crowd: We just pushed v0.2.2 to NPM.

We've officially moved past the 'honor system' to a hardened x402 on-chain settlement logic. The server now performs real-time verification of USDC transfers on Base before granting credits to solve the gas-to-value friction ($1.00 for 1,000 frames).

500 Frame Sponsorship: We've baked the 500-frame sponsorship directly into the build. If you run npx -y glazyrviz, the Big Iron node handles the first 500 frames on the house—no wallet needed for the initial benchmark.

Detailed build notes on the ThinLTO/CFI hardening and the Viz-patching logic: https://glazyrviz.blogspot.com/2026/02/inside-zero-copy-engi...

For anyone wanting the deep-dive on the binary hardening and Viz sub-system patches I mentioned, I just offloaded the full technical analysis here:https://glazyrviz.blogspot.com/2026/02/inside-zero-copy-engi.... Also, I’ve officially bumped the sponsorship pool to 500 free frames for everyone testing npx -y glazyrviz today.

Actually, for the early beta testers here on HN, I’m upping the sponsorship to 500 free frames. That should give you plenty of headroom to stress-test the Zero-Copy pipeline on some heavier sites.

No ETH needed—just run the npx command and the Paymaster handles the rest on Base

I’ve been obsessed with the "last mile" bottleneck in AI agents. Traditional web automation relies on CDP (Chrome DevTools Protocol) to serialize the DOM or take screenshots. This "capture-encode-transmit" loop is computationally ruinous and introduces latencies that often exceed the reasoning time of the underlying LLM.

I built Glazyr Viz (originally Neural Chromium) by forking Chromium to integrate the agent directly into the browser’s compositor subsystem (Viz).

Key Technicals:

Zero-Copy Vision: We establish a Shared Memory (SHM) segment between the Viz process and the agent using shm_open. This allows the agent to access raw frame buffers with sub-16ms latency—essentially syncing with the 60Hz refresh rate.

Hardening (ThinLTO/CFI): Security is usually a performance tax, but we achieved a "Performance Crossover." By coupling Control Flow Integrity (CFI) with ThinLTO (LLVM 19), we achieved a 40% speed increase in JS execution and an 85.8% reduction in P99 latency jitter.

Context Density: Instead of raw markdown, we deliver structured context via a vision.json schema. On our "Big Iron" (GCE n2-standard-8) cluster, we’re hitting 177 TPS.

Economic Layer (x402): I wanted a way for machines to pay machines. We use the x402 (Payment Required) status on Base (USDC). It’s gasless via EIP-7702 delegation, so agents can pay-per-frame without holding ETH.

Why this matters: If you're building agents for high-frequency tasks (live trading, dynamic UI auditing, or real-time navigation), standard scrapers are too brittle and slow. Glazyr Viz treats the browser as a high-speed sensory organ rather than a slow external portal.

Try it: I’ve published the orchestration layer to NPM. You can spin up a local node and hook your agent into it immediately: npx -y glazyrviz

I’d love to hear from anyone working on low-level rendering or agentic infrastructure. Happy to dive into the shm_open implementation or the ThinLTO pass details in the comments.