I made a browser trainer for blindfold chess practice
I wanted exercises for practising without seeing the pieces. Blindfold Chess covers coordinates, movement, notation, and position updates.
The public version needs no account and saves progress locally.
Play: ceasar369.github.io
Source and local edition (MIT): Ceasar369/blindfold-chess
The browser interface uses React. The downloaded edition can save completed rounds as files.
I’d appreciate feedback on the drills and local setup.
2 points · 1 comment
Blindfold Chess — a free browser trainer for visualisation practice
Playable Link:
ceasar369.github.io
Platform:
Web browser
Description:
I wanted to practise chess without always seeing the pieces. Blindfold Chess breaks that practice into separate exercises. You can work on coordinates, piece movement, and notation. Other chapters cover remembering positions and updating them after moves. There is also a blind-play chapter.
You can start with board coordinates before trying harder exercises. For example, one exercise asks you to find a square. The early version includes names along the board’s edge. Later exercises give you less help.
It runs in your browser without an account. Your progress stays on your device. The source and a downloadable local edition are available too. I’d appreciate feedback on confusing instructions or awkward controls. I’m especially interested in where the difficulty jumps too quickly.
Free to Play Status:
Free to play
Involvement:
I made this for my own practice. I chose the exercises and behaviour, using Claude for implementation. The code is public under the MIT licence:
Ceasar369/blindfold-chess
1 point · 1 comment
Claude Behavior Layer — the rules and review pages I use with Claude Code
This is how I set up Claude Code. I wanted short answers and decisions I could review easily.
It combines writing rules, reusable skills, and local browser pages. Claude puts questions or prototype choices there for me. I review them, then send my answers back.
The complete setup currently uses macOS, iTerm2, and Python. Individual rules and skills can be reused separately. It’s free, with an MIT licence.
I also used it while building my Blindfold Chess trainer. Both repositories are public if you want a concrete example.
Chess example: Ceasar369/blindfold-chess
I’m sharing my own setup in case it helps someone. What would make the setup easier to try?
1 point
I made a blindfold chess trainer for my own practice
I wanted a way to practise without seeing the pieces. So I made Blindfold Chess.
It includes coordinate, movement, notation, and position exercises. It runs in your browser without an account. Progress stays on your device.
Play: ceasar369.github.io
Code and local download: Ceasar369/blindfold-chess
I used my Claude Code setup to build it. That setup is also public: Ceasar369/claude-behavior-layer
I’d like feedback on exercises that feel awkward or unclear.
3 points · 10 comments
The Claude Code setup I use for my projects
I’ve put my Claude Code setup on GitHub. It works well for me. I’m sharing it in case parts of it are useful.
The main pieces are rules, skills, and local decision pages. The rules ask for short answers and clear decisions. The skills cover planning, coding, prototypes, checks, and handovers. They also coordinate separate Claude Code sessions.
For example, a session proposes two implementation choices. I review both on a local page. I can agree or add feedback before it continues. A handover saves written context for a fresh session.
I also tried the same setup with Blindfold Chess. It’s a trainer I wanted for my own practice. It has exercises for playing without seeing the pieces. The same setup worked well on that separate project.
The full setup uses macOS, iTerm2, and Claude Code. You can also take just the rules or skills. The code is MIT licensed.
Claude setup: Ceasar369/claude-behavior-layer
Chess: ceasar369.github.io
The screenshots use a fictional reading-list project. They show the real pages with prepared demo content.
Demo: reviewing where the reading list saves articles.
Demo: assistant plan with clarification, prototypes, implementation, and checks.
Demo: comparing three layouts for the same reading list.
Demo: opening one prototype before choosing or giving feedback.
1 point · 7 comments
Mentions
I made a browser trainer for blindfold chess practice I wanted exercises for practising without seeing the pieces. Blindfold Chess covers coordinates, movement, notation, and position updates. The public version needs no account and saves progress locally. Play: ceasar369.github.io Source and local edition (MIT): Ceasar369/blindfold-chess The browser interface uses React. The downloaded edition can save completed rounds as files. I’d appreciate feedback on the drills and local setup.
Blindfold Chess — a free browser trainer for visualisation practice Playable Link: ceasar369.github.io Platform: Web browser Description: I wanted to practise chess without always seeing the pieces. Blindfold Chess breaks that practice into separate exercises. You can work on coordinates, piece movement, and notation. Other chapters cover remembering positions and updating them after moves. There is also a blind-play chapter. You can start with board coordinates before trying harder exercises. For example, one exercise asks you to find a square. The early version includes names along the board’s edge. Later exercises give you less help. It runs in your browser without an account. Your progress stays on your device. The source and a downloadable local edition are available too. I’d appreciate feedback on confusing instructions or awkward controls. I’m especially interested in where the difficulty jumps too quickly. Free to Play Status: Free to play Involvement: I made this for my own practice. I chose the exercises and behaviour, using Claude for implementation. The code is public under the MIT licence: Ceasar369/blindfold-chess
Claude Behavior Layer — the rules and review pages I use with Claude Code This is how I set up Claude Code. I wanted short answers and decisions I could review easily. It combines writing rules, reusable skills, and local browser pages. Claude puts questions or prototype choices there for me. I review them, then send my answers back. The complete setup currently uses macOS, iTerm2, and Python. Individual rules and skills can be reused separately. It’s free, with an MIT licence. I also used it while building my Blindfold Chess trainer. Both repositories are public if you want a concrete example. Chess example: Ceasar369/blindfold-chess I’m sharing my own setup in case it helps someone. What would make the setup easier to try?
I made a blindfold chess trainer for my own practice I wanted a way to practise without seeing the pieces. So I made Blindfold Chess. It includes coordinate, movement, notation, and position exercises. It runs in your browser without an account. Progress stays on your device. Play: ceasar369.github.io Code and local download: Ceasar369/blindfold-chess I used my Claude Code setup to build it. That setup is also public: Ceasar369/claude-behavior-layer I’d like feedback on exercises that feel awkward or unclear.
The Claude Code setup I use for my projects I’ve put my Claude Code setup on GitHub. It works well for me. I’m sharing it in case parts of it are useful. The main pieces are rules, skills, and local decision pages. The rules ask for short answers and clear decisions. The skills cover planning, coding, prototypes, checks, and handovers. They also coordinate separate Claude Code sessions. For example, a session proposes two implementation choices. I review both on a local page. I can agree or add feedback before it continues. A handover saves written context for a fresh session. I also tried the same setup with Blindfold Chess. It’s a trainer I wanted for my own practice. It has exercises for playing without seeing the pieces. The same setup worked well on that separate project. The full setup uses macOS, iTerm2, and Claude Code. You can also take just the rules or skills. The code is MIT licensed. Claude setup: Ceasar369/claude-behavior-layer Chess: ceasar369.github.io The screenshots use a fictional reading-list project. They show the real pages with prepared demo content. Demo: reviewing where the reading list saves articles. Demo: assistant plan with clarification, prototypes, implementation, and checks. Demo: comparing three layouts for the same reading list. Demo: opening one prototype before choosing or giving feedback.