Feiyu's Technical Blog
Notes on software, systems, and things I am learning.
Agents & Automation
Local models, tool-using agents, and practical information-extraction workflows.
Automating 1Point3Acres Daily Check-In and Quiz Answering with a Local LLM
A Windows app runs around 9:00 AM, uses signed-in Chrome to complete a forum check-in, and starts a local LLM only when a quiz needs an answer.
[2/2] How I Made a Codex-to-ChatGPT Bridge Verifiable
A narrow local agent reuses my signed-in ChatGPT conversation, submits each request once, and verifies short, PDF, and 10,000-character replies without resending.
[1/2] (Deprecated) How I Let Codex Talk to ChatGPT Through My Signed-In Chrome
My first Codex-to-ChatGPT bridge let Codex control Chrome directly. It proved the idea worked, but the design was too fragile to keep.
Local LLM + Agent for Web Information Extraction
A local agent architecture for semantic web extraction, browser automation, bounded model context, and durable task state.
Game Perception — Olden Era
Computer-vision experiments that turn arena gameplay screenshots into structured game state.
真实游戏截图能否改进只用 Catalog 训练的游戏 UI 图标分类器?
一次固定随机种子的受控开发实验:比较 Catalog-only 与 verified gameplay supervision,并测试 learned validity 是否比 softmax confidence 更会拒绝无效 crop。
从一次模型退化审计到可复现实验:重构游戏 UI 图标分类器
我们冻结 2,136 个可追溯 crop,用五个随机种子比较 catalog-only、gameplay adaptation 与 learned rejection,并逐样本审计 B04 的结果和边界。
Training a Game UI Icon Detector Using Only the Icon Catalog
Using 390 catalog images—and no gameplay screenshots for weight updates—we test full-screen hero and skill detection.
Full-Image Icon Search with Sliding Windows and Visual Embeddings
In this seven-frame experiment, the search often identifies the right hero or skill, but its highest-scoring box is not always the best crop.