The 2.7KB Zig WASM binary is the scoring engine that runs on every request at Cloudflare's edge. The globe visualizes where those requests land. Two layers — compute at the edge, visualization in the browser.
MCPaaS serves persistent AI context via the Model Context Protocol. A namepoint (mcpaas.live/yourhandle) gives your AI instant project context — no re-explaining every session. Works with Claude, Gemini, Cursor, any MCP client.
Exactly right. 2.7KB works because it's pure computation — slot counting, no allocator, no stdlib, no WASI. The moment you need I/O it balloons. This use case fits a glove
Fair point — globe.gl (Three.js) handles the 3D rendering client-side.
The 2.7KB WASM is the server-side scoring engine — Zig-compiled, runs on every
request at the Cloudflare edge. The globe visualizes where those executions happen.
Two separate layers: WASM at the edge, JS in the browser.
ETH Zurich tested this: LLM-generated prose context = -3% performance, +20% cost. Even human-written = +4% at +19% cost. The problem is prose bloat. Structured formats avoid that by design.
https://arxiv.org/abs/2602.11988
ln -s makes all four files identical. Whichever format you write it in, the other three get the wrong structure. This generates each in its native format.
Totally — git handles syncing files. The problem is these four files have different formats and conventions. Same project context, four dialects. That's why I wrote bi-sync --all: one YAML source, four native outputs.
Well said. And it's potentially a 7% swing when you think about it — +4% with good human-written context vs. -3% with LLM-generated noise. That's a significant delta from just the quality of the information.
The real value is exactly what you described: the tribal knowledge, the "we tried X and it broke because Y", the constraints that live in someone's head and nowhere in the code. LLM-generated files miss this because the LLM is just restating what it can already see. Of course that doesn't help.
This paper validates what we've been building toward. The core issue isn't the idea of context files — it's that prose is the wrong format for structured facts.
AI crushes structured data like package.json but struggles with free-form markdown. Two developers describe the same repo completely differently. There's no schema, no validation, no scoring.
Our paper on CERN's Zenodo proposes FAF — a structured YAML format (IANA-registered as application/vnd.faf+yaml) that replaces prose with validated fields. One .faf file generates native outputs for CLAUDE.md, AGENTS.md, .cursorrules, and GEMINI.md. The instruction files stay — they just sit on top of a structured foundation instead of floating independently.