How Might a Modern Lighthouse Work With LLMs?

What it is — A proposal for a Lighthouse-style auditing tool where checks are plain-language descriptions of good practice that an LLM judges from screenshots and traces, instead of hand-coded rules.

Key points

• Combines a spec of good practices, machine-readable fix guidance, and evidence primitives (screenshots, heap snapshots, perf traces) as the model’s “senses.”
• Runs a continuous find-fix-test loop rather than a one-shot audit, and reports fixing several real site issues this way (still with human review).
• Because tests are descriptions of outcomes rather than code, anyone who can describe “good” can author a check — widening coverage into fuzzier, subjective quality dimensions.