Feiyu's Technical Blog

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.

  1. A 1Point3Acres Assistant with a Local LLM and Codex Review

    Why an LLM suggestion, permission to submit, and a verified website result need separate checks.

  2. [2/2] How I Made a Codex-to-ChatGPT Bridge Verifiable

    A local bridge sends one authorized request and brings back a checked answer. Its failures explain what message identity, durable state and realistic tests must get right.

  3. [1/2] (Deprecated) How I Let Codex Talk to ChatGPT Through My Signed-In Chrome

    The first prototype made the round trip work. Its failures taught me why browser mechanics need a smaller, testable home.

  4. 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

Three ways to turn Olden Era Arena screenshots into structured game state, organized by how the system finds the icons—not by publication date.

ApproachVisible text · ordered workflow

OCR Scene Understanding

Read visible text to identify the current screen, then keep workflow progress separate from current-frame evidence.

  1. Text and position
  2. Current screen
  3. Phase context

ApproachKnown geometry · nearly fixed camera

Calibrated Grid Registration

Use the battlefield’s regular hexagonal geometry and a saved calibration to infer cell positions, then refine their alignment from screenshot edges.

  1. Calibrated grid
  2. Image edge scoring
  3. Small alignment correction

ApproachKnown position

Fixed-Crop Classification & Rejection

Assume each target icon appears at a known UI location. With localization removed, perception becomes a classification-and-rejection problem.

  1. Known UI slot
  2. Fixed crop
  3. Classify + reject

ApproachUnknown position · learned localization

Learned Object Detection

Do not assume the icon location is known. A learned detector proposes boxes; a separate catalog matcher suggests identities. The current evidence is a small selection-screen pilot.

  1. Full screenshot
  2. Learned detector
  3. Boxes + identities

ApproachKnown layout · readable titles

Printed-Name Recognition

Read the name printed beside an icon, resolve bilingual aliases, and propose an identity for human review. Readable text does not prove crop validity.

  1. Title region
  2. Local OCR
  3. Canonical name + review

ApproachUnknown position · code-generated candidates

Sliding-Window Icon Detection

Search the full screenshot with many candidate crops, encode each crop into visual features, and match them against the known icon catalog.

  1. Full screenshot
  2. Candidate crops
  3. Catalog matching