chess-best-move

Analyze chessboard images to detect pieces and compute best moves.

134|21|Updated Nov 12, 2025
One-click install
npx skills add https://github.com/letta-ai/skills --skill chess-best-move
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chess-best-move
Source: https://github.com/letta-ai/skills/tree/main/ai/benchmarks/letta/terminal-bench-2/trajectory-feedback/chess-best-move
Command: npx skills add https://github.com/letta-ai/skills --skill chess-best-move

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill guides image-based chess analysis to determine the best move(s) from a given board position, emphasizing robust board recognition over pattern matching.

Core Features & Use Cases

  • Systematic board detection and orientation
  • Piece identification across all squares and construction of a valid position
  • Move calculation using a chess engine or library
  • Output formatting for one or more winning moves

Quick Start

Provide an image of a chess position to the AI and retrieve the best move(s) in standard notation.

Frequently Asked Questions about chess-best-move

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a chess board image to find the best move?

Provide a chess board image to the Skill, which detects the board orientation, identifies all piece positions, constructs a valid position, and uses a chess engine to calculate and return the best move in standard notation.

Can this detect chess pieces accurately from photographs or screenshots?

Yes, the Skill performs systematic board detection and per-square piece identification across all squares to construct a valid position, handling various image sources rather than relying on pattern matching alone.

What happens if the chess position in the image is illegal or invalid?

The Skill includes position legality verification as part of its core processing, ensuring only valid chess positions are analyzed before move calculation and output.

Can it calculate mate sequences or just single best moves?

The Skill computes winning moves and mate-in-N sequences, returning one or more best moves depending on the position analysis from the chess engine.

What image formats work with board detection and piece identification?

The Skill accepts chess board images for analysis; it performs systematic board detection and orientation to extract piece positions, though specific supported formats depend on the image-analysis component implementation.

Do I need chess knowledge to interpret the output moves?

Output is formatted in standard chess notation, which requires familiarity with algebraic notation to interpret; the Skill handles all board recognition and engine computation.