task_puzzles__cube_net__opposite_face_label

Summary

1. Domain: puzzles
2. Scene id: cube_net
3. Task id: task_puzzles__cube_net__opposite_face_label
4. Objective contract: choose the option matching the face opposite the marked face after folding.

Program Contract

Program: select_option(folded_cube_net.opposite_face, reference=marked_face); scene=cube_net; scope=opposite_face_label

Candidate set: the visible cube-net faces, face colors/labels, marked face or edge, and labeled candidate options inside the opposite_face_label objective scope. Operands: visible scene state and prompt-bound operands named by folded_cube_net, opposite_face, reference, marked_face, cube_net, opposite_face_label. Operation: evaluate select_option over the candidate set using the visible states, constraints, transforms, comparisons, counts, paths, or option-selection rules encoded in the program expression; generation enforces a unique final answer. Output binding: answer uses the option_letter schema; generation binds a unique final answer. Annotation witnesses: annotation uses the bbox schema; the prompt/annotation contract defines the minimal visual witnesses. Query ids: single.

Reasoning Operations

Families: transformation

Query Contract

1. Public query id: single
2. The task always asks for the folded-cube face opposite the marked reference face.
3. Net rotation, style, face labels, and option order are render/generation axes, not query ids.

Answer And Annotation

1. Answer type: option_letter
2. Annotation schema: bbox
3. Annotation box is the image-pixel box for the selected option card.
4. The marked reference face is already visible in the image and is not part of the annotation.

Implementation

1. Registered class: trace_tasks.tasks.puzzles.cube_net.opposite_face_label.PuzzlesCubeNetOppositeFaceLabelTask
2. Prompt bundle: src/trace_tasks/resources/prompts/puzzles/cube_net/puzzles_cube_net_v1.json
3. Scene config: src/trace_tasks/resources/configs/domains/puzzles/cube_net.yaml