the real engine, compiled to WASM zero ML · zero network · zero server sub-millisecond decisions

Decide something, right here.

This page embeds the same deterministic engine the servers run — rules, candidate narrowing, lexical scoring, calibrated confidence, honest abstention — compiled to WebAssembly and executed entirely in your browser tab. Pick a scenario, edit the JSON, press Decide. The status chips are measured live, not asserted.

Engine
loading…
WASM
loading…
Network
measuring…
Server
none — this page is static files
LLM
none — zero-ML stack (rungs 1–4)
Playground

The full request → decision → trace loop

A request carries state, questions, and policy. The engine returns a decision per question with a multi-dimensional confidence report, execution metrics, and a deterministic trace. Refusals are rendered as the outcomes they are — the gate working, not an error.

Anything you type stays in this tab — there is no server to send it to. Invalid JSON or invalid requests return the same canonical error codes the HTTP API returns.

loading

Booting the WASM engine…

What this is — and what it is not

  • Real: the WASM module is built from this repo's opencodifier-wasm crate — the identical IR, rule engine, narrowing, lexical scorer, cache, and calibration code the CLI and HTTP server run.
  • Real: the zero-ML stack only. Rungs 5–7 (embedding, decision model, verifier) need backends this tab deliberately does not ship; a graph that needs one returns engine.missing_backend instead of pretending.
  • Not here: network calls, telemetry, or model downloads. The Network chip counts cross-origin requests from the Performance API — if it ever reads anything but zero, the page is broken, and the chip will say so.

Prefer a terminal? The same decisions from cargo run --bin opencodifier-cli -- decide, or serve the HTTP API and hit the documented endpoints.