Browser harness contract and blind spots

Lists the static contract enforced by the no-crate harness and separates it from actual browser execution, accessibility, rendering, and inference verification.

Experimental
Last verified
2026-06-25 00:00 UTC
Updated
Reading time
2 minutes

Lists the static contract enforced by the no-crate harness and separates it from actual browser execution, accessibility, rendering, and inference verification.

Implementation evidence: this topic is grounded in the reviewed GGUF.MiRust.com source snapshot. It documents observed code and artifacts without claiming broad deployment, model quality, or production readiness.

Architecture topic: this page does not claim that the WordPress website implements or executes the described runtime behavior.

Static checks

The harness requires 37 DOM IDs, visible privacy and product markers, local stylesheet/script routes, required JavaScript ABI call markers, 13 Rust exports, eight model routes, adjacent manifests with 14 keys, artifact presence, expected quantization labels, and absence of remote/CDN/analytics markers.

Optional HTTP checks

With --http, it probes a loopback app URL and required local routes with a small TCP client.

Blind spots

String presence does not prove JavaScript control flow, CSS layout, keyboard operation, browser API compatibility, model correctness, screen-reader behavior, or performance. Headless browser smoke and human review remain separate gates.

Scope

This starter page defines the questions, boundaries, evidence, and failure modes that should be recorded before a capability is presented as supported.

Engineering considerations

  • Identify the source, version, target environment, and owner.
  • Separate observed values from estimates and externally reported values.
  • Record trade-offs, unsupported cases, and fallback behavior.
  • Link performance statements to a compatible benchmark methodology.

Verification questions

  • What exact artifact, revision, backend, and environment were reviewed?
  • Which assumptions could change the result?
  • Which data should be retained so another engineer can reproduce the conclusion?