jev-chess
npm library for chess with Jev: natural-language move intents to legal moves, parallel move evaluation and historical persona opponents.
Overview
jev-chess by Hemanth is a TypeScript library that uses Jev for chess tasks while a chess engine handles legality. resolveIntent maps a natural-language instruction, such as develop knight to attack centre or castle kingside to safety, to a verified legal move via a Choice over legal moves, returning alternatives instead of guessing when confidence falls below a threshold. evaluateMove assesses a move in parallel for tactical sharpness (a Score from 0 to 3), strategic theme (a Choice such as pawn break, tactical strike or prophylaxis) and king risk.
selectMove weighs candidate moves against persona archetypes, offering historical opponents such as Tal, Capablanca and Petrosian with a rationale for each move, and the list summaries mention game classification as well. It is installed from npm and aimed at developers building chess apps that want natural-language interaction or commentary-like evaluation without generating coordinates.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which legal move matches this natural-language intent? | — |
| Score | How tactically sharp is this move (0-3)? | — |
| Choice | What is the strategic theme of this move? | pawn_break · tactical_strike · prophylaxis · ... |
How it works
- Engine lists legal moves
- resolveIntent maps natural language to a legal move
- evaluateMove scores sharpness, theme and king risk in parallel
- Persona engine selects opponent replies with rationales
Strengths
- Never produces illegal moves
- Returns alternatives when unsure
- Parallel evaluation in one call
Caveats
- Persona play is not engine-strength chess
- No licence stated in the README excerpt
Getting started
`npm install jev-chess`, then use ChessEngine, MoveResolver, MoveEvaluator and PersonaEngine.At a glance
- Type
- Library
- Status
- beta
- Language
- TypeScript
- Runtime
- library
- Has code
- yes
Use cases and integrations
How curated lists describe it
Chess moves, evaluations, persona opponents, and game classification using TypeSafe AI System One models. Resolves natural language move intents into legal moves, evaluates positional sharpness and king risk in parallel, and powers historical persona opponents (Tal, Capablanca, Petrosian).
logicrw/awesome-jev-projectsChess moves, evaluations, persona opponents, and game classification with TypeSafe AI System One.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev decide?
Which legal move matches an instruction, how to characterise a move, and which move a persona would choose.
Can it make illegal moves?
No, it chooses among verified legal moves.
Which personas exist?
Examples include Tal, Capablanca and Petrosian.
Related projects
Post evaluating Jev for mapping free-text player intent into typed server actions in a tabletop MMORPG.
DiscussiondiscussionSource for browser chess and Chess960 where code computes each legal move's facts and Jev picks one per turn as a typed Choice.
ProjectproductionTypeScriptBrowser 1v1 sniper arena where Jev reads match state about 9 times a second and chooses movement, aim, ADS, firing and jumping.
ServicebetaTypeScriptInstrumented 2048 web lab where every move is a Jev Choice over four directions, with probabilities, latency and cost shown live.
ProjectbetaTypeScriptFifteen live arcade games where Jev and LLMs play each other, led by WikiRace in which Jev picks only real links.
ProjectbetaTypeScriptReal-time bot battles where you write a fighter's strategy in plain English and a System One model pilots it several times a second.
ProjectbetaTypeScript