task_symbolic__life_automaton__life_future_grid_label

Summary

1. Domain: symbolic
2. Scene: life_automaton
3. Task id: task_symbolic__life_automaton__life_future_grid_label
4. Goal: apply a cellular-life neighbor rule to the grid marked START and choose the option showing the future grid.

Program Contract

Program: life_automaton.future_grid_option_label(scene=life_automaton, scope=marked_start_grid_plus_option_set, grid_size=3x3|4x4|5x5, update_steps=one|two, output=option_letter)

Candidate set: the visible future-grid option panels. Operands: the START grid state, grid size, life update rule, and requested one-step or two-step horizon. Operation: apply the life rule for the requested number of updates and select the unique option grid matching the result. Output binding: answer is the selected option letter. Annotation witnesses: a bbox_map with source_grid and selected_option roles. Query ids: one_step_future_grid, two_step_future_grid.

Reasoning Operations

Families: state_update

Contract

1. Query ids: one_step_future_grid|two_step_future_grid
2. Answer type: option_letter
3. Annotation schema: bbox_map
4. Annotation target: role-keyed bboxes for source_grid and selected_option
5. Scene variants: clean_grid|lab_panel|notebook_grid
6. Prompt bundle: symbolic_life_automaton_v1
7. Source file: src/trace_tasks/tasks/symbolic/life_automaton/life_future_grid_label.py
8. Render metadata: records panel style, role-aware font family, unit-size jitter, and annotation-safe layout jitter before annotation projection.
9. Visual contract: source grid is visibly marked START; source grid and option grids use the same rendered cell scale.
10. Scene-local variation: records life_board.board_style, life_board.cell_palette_id, resolved RGBs, and alive/dead/marker contrast checks.

Annotation Contract

The annotation is a bbox_map because the task has two distinct witness roles: the source grid that must be updated and the selected visual option panel.