Chess Stalker v3.0 - Opponent Analysis Tool with In-Browser Stockfish

[Rujas]

Hi everyone,

I’m David, a CS student and solo developer from Spain. I wanted to share a project I’ve been working on: Chess Stalker
What it does:
Chess Stalker analyzes any player’s games (Lichess, Chess.com, or FIDE OTB) and generates a preparation report:
∙ Opening repertoire with exploitable weaknesses
∙ Psychological profile (tilt tendency, time management, resignation patterns)
∙ Traps they use and fall for
∙ Statistical breakdown of their play

Technical highlights:
∙ In-browser Stockfish 16 WASM - Full analysis runs client-side, no server load
∙ Twin Bot - A bot that plays the opponent’s actual opening repertoire, then transitions to Stockfish at adjustable ELO
∙ 11M+ FIDE OTB games indexed with Stockfish-analyzed positions
∙ React frontend, Python/Flask backend, PostgreSQL for the FIDE database

Why I’m posting here:
I’d love feedback from this community, especially on:
∙ The Stockfish WASM integration
∙ Ideas for improving the “Twin Bot” transition from book moves to engine play
∙ Any suggestions from people who understand chess engines better than I do
It’s free to use. 65.000+ analyses completed only for online accounts so far.

Happy to answer any technical questions!
David

[Ipmanchess]

https://www.chessstalker.com/

[Rujas]

https://www.chessstalker.com/

Thanks for sharing the link. I just forgotten
Pls, feel free with your feedback! Even if you miss any feature - all ideas are welcome

[pohl4711]

https://www.chessstalker.com/

Thanks for sharing the link. I just forgotten
Pls, feel free with your feedback! Even if you miss any feature - all ideas are welcome

Did you consider to use the superagressive Engine Patricia? Analyzing human games with Patricia delivers a bunch of crazy, aggressive moves/ideas, Stockfish will not even consider. So, Patricia is perfect as a second engine and as an addition to Stockfish. Patricia has a full working Multi-PV-Analyzing mode. And because her neural-net is very small, Patricia has high nps and can be used well with not-so-much hardware power. And the binary of Patricia is also very small, less than 5 MByte, including nets.
The latest official release (Patricia 5) is on a strength-level with Stockfish 10/11. So, still superhuman-level. But Patricia is winning more than 50% of her games with a sacrifice - Stockfish 18 can only dream of this.

All informations and download can be found on my website:
https://www.sp-cc.de/patricia_eas_engine.htm

Direct download of Patricia 5, (all binaries needed, made by Jim Ablett):
https://drive.proton.me/urls/YTAHHXANQ4#HJ9O7qoQZIxH

"I am checking games with Patricia engine. It's totally crazy! I haven't seen any engine play like this" - GM Vidit Gujrathi (world ranking #26, Indian Chess Olympiad gold medalist in 2024), 2025/07/27 on Lc0-discord

[Rujas]

https://www.chessstalker.com/

Thanks for sharing the link. I just forgotten
Pls, feel free with your feedback! Even if you miss any feature - all ideas are welcome

Did you consider to use the superagressive Engine Patricia? Analyzing human games with Patricia delivers a bunch of crazy, aggressive moves/ideas, Stockfish will not even consider. So, Patricia is perfect as a second engine and as an addition to Stockfish. Patricia has a full working Multi-PV-Analyzing mode. And because her neural-net is very small, Patricia has high nps and can be used well with not-so-much hardware power. And the binary of Patricia is also very small, less than 5 MByte, including nets.
The latest official release (Patricia 5) is on a strength-level with Stockfish 10/11. So, still superhuman-level. But Patricia is winning more than 50% of her games with a sacrifice - Stockfish 18 can only dream of this.

All informations and download can be found on my website:
https://www.sp-cc.de/patricia_eas_engine.htm

Direct download of Patricia 5, (all binaries needed, made by Jim Ablett):
https://drive.proton.me/urls/YTAHHXANQ4#HJ9O7qoQZIxH

"I am checking games with Patricia engine. It's totally crazy! I haven't seen any engine play like this" - GM Vidit Gujrathi (world ranking #26, Indian Chess Olympiad gold medalist in 2024), 2025/07/27 on Lc0-discord

Wow, I didn’t know it. I will explore it.
Are you involved on that project? I would like to talk with you in case of

Thanks

[pohl4711]

Wow, I didn’t know it. I will explore it.
Are you involved on that project? I would like to talk with you in case of

Thanks

Directly? No. Patricia is done by Adam Kulju (nickname in talkchess: Whiskers)
https://github.com/Adam-Kulju/Patricia

But I had the idea and made the EAS-Tool. The EAS-Tool made it possible to measure aggressiveness in engine-games and without it, Patricia would not exist.

Adam Kulju:
"The metric that Patricia's aggression claims are based off of is Stefan Pohl's EAS-Tool, which is the most well known and well regarded tool for determining the aggressiveness of chess engines. It looks at a combination of factors, such as sacrifice rate, short win rate, and unnecessary draw rate, and outputs a score that captures how "exciting" an engine tends to play.
His EAS Tool works wonderfully, makes properly and objectively testing for increase aggression possible, and is the measure by which Patricia development progressed. He was also very invested in Patricia's development and explained many features of the EAS tool to me so that I had a better understanding of what was a sacrifice and what wasn't. He has put a lot of time into quantifying style, and I am happy to have Patricia be a proof-of concept and culmination of his ideas."