BRIGHTON LABWORKS / 02

Making things,
to find things out.

A running index of things I’m building, studying, and trying to understand across computing.

01

Current project

DIARYBENCH

Built from my private Chinese diary, DiaryBench tests fine-grained emotional analysis across a long sequence of diary entries, rather than merely rewarding a plausible-sounding analysis.

Diagram showing DiaryBench's Chinese diary inputs, author persona constraints, model evaluation, and alignment checks
Open the benchmark board
02

Other work

Agent Skills

A small set of review, vision, frontend, and orchestration skills that I use while making things with Agents.

Download the bundle 4 skills
  1. 01
    gpt-web-review

    Deep review and external critique through ChatGPT web.

  2. 02
    mimo-vision

    Visual inspection for screenshots, frames, charts, and OCR; configure a local OpenCode credential before use.

  3. 03
    frontend-design

    Distinctive interface direction with a guard against template aesthetics.

  4. 04
    orchestration

    Research, build, and critique loops for non-trivial work.