task_symbolic__music_staff__transposed_pitch_pair_count

1) Identity

1. Domain: symbolic
2. Scene id: music_staff
3. Task id: task_symbolic__music_staff__transposed_pitch_pair_count
4. Objective contract: transposed_pitch_pair_count

Program Contract

Program: music_staff.transposed_pitch_pair_count(scene=music_staff, scope=four_numbered_note_pairs, predicate=upward_interval_match, output=integer)

Candidate set: the four visible numbered note-pair ranges. Operands: the two notes in each pair and the requested upward interval predicate. Operation: compute the upward interval for each note pair and count pairs matching the requested interval. Output binding: answer is the matching pair count as an integer. Annotation witnesses: a homogeneous bbox_set of numbered note-pair range boxes that satisfy the requested interval. Query ids: single.

Reasoning Operations

Families: filtering, counting, transformation

2) Scene + task contract

1. Entities/relations: a rendered music-staff notation panel with marked notes, chords, measure ranges, option cards, or key signatures depending on the task objective.
2. Supported query_id values: single
3. answer_gt.type: integer
4. Default annotation_gt.type: bbox_set
5. Annotation schema: bbox_set
6. Annotation witness policy: annotation marks numbered note-pair range boxes that satisfy the requested upward interval; non-target notes, nonmatching ranges, staff lines, and decorative background are not prompt-facing annotation.

Query id	User-facing operation
single	The prompt asks the one stable objective contract for this task.

3) Prompt contract

1. prompt_bundle_id: symbolic_music_staff_v1
2. scene_key: music_staff
3. task_key: music_transposed_pitch_pair_count
4. Query keys: internal branch key from the generated trace; public single-operation tasks expose query_id=single after registry normalization.
5. Output modes: answer_only, answer_and_annotation

4) Determinism + constraints

1. Deterministic generation and rendering from instance_seed.
2. Answer and annotation are bound from the same finalized notation scene and projected bboxes.
3. Count tasks construct the requested answer count before rendering; label tasks construct a unique target label by scene state.
4. No semantic constraints are auto-relaxed after sampling.

5) Source files

1. Task source: src/trace_tasks/tasks/symbolic/music_staff/transposed_pitch_pair_count.py
2. Scene shared package: src/trace_tasks/tasks/symbolic/music_staff/shared/
3. Config: src/trace_tasks/resources/configs/domains/symbolic/music_staff.yaml
4. Prompt asset: src/trace_tasks/resources/prompts/symbolic/music_staff/symbolic_music_staff_v1.json
5. Focused tests: tests/test_symbolic_notation_tasks.py