task_graph__graph_options__contained_subgraph_label

Summary

1. Domain: graph
2. Scene id: graph_options
3. Task id: task_graph__graph_options__contained_subgraph_label
4. Objective: select the option graph contained as a labeled subgraph in the Target Graph.
5. Implementation: src/trace_tasks/tasks/graph/graph_options/contained_subgraph_label.py.

Query IDs

1. Supported query_id values: single.
2. Internal prompt key: contained_subgraph_label.
3. Public sampling is at the task-id level.

Taxonomy Contract

1. Program contract: compare the Target Graph with four visual option graphs and choose the option whose node labels and edges are all present in the Target Graph.
2. Answer schema: option_letter.
3. Annotation schema: bbox.
4. Edge direction, target node count, subgraph node count, correct option slot, style, font, background, and layout jitter are generation/render metadata, not public query branches.

Program Contract

Program: select(option_graph where labeled_edges(option_graph) subset_of labeled_edges(target_graph)); output=option_letter; annotation=bbox(selected_option_panel); scene=graph_options; scope=contained_subgraph_label

Candidate set: the visible graph, tree, network, route, matrix, table, node, edge, label, weight, path, and option elements inside the contained_subgraph_label objective scope. Operands: visible scene state and prompt-bound operands named by option_graph, where, labeled_edges, subset_of, target_graph, selected_option_panel, graph_options, contained_subgraph_label. Operation: evaluate select over the candidate set using the visible graph structure, labels, weights, directions, reachability, paths, connectivity, or option-selection constraints 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: topology, matching

Answer And Annotation

1. Answer type: option_letter.
2. Annotation type: bbox.
3. Annotation marks the selected visual option panel as one [x0,y0,x1,y1] pixel-space box.
4. The selected option panel and answer letter come from the same execution trace.

Rendering Contract

1. The scene shows one Target Graph above four visual graph-option panels.
2. Directed samples use arrows and require exact direction matching.
3. Visual style, fonts, panel treatment, layout jitter, and post-render noise are non-semantic and recorded in trace metadata.
4. Annotation projection is computed after final layout and style placement.

Prompt Contract

1. Prompt text comes from src/trace_tasks/resources/prompts/graph/graph_options/graph_options_v1.json.
2. scene_key: graph_options.
3. task_key: structure_match_label_query.
4. query_key: contained_subgraph_label.
5. Answer mode emits {"answer": ...}.
6. Answer-and-annotation mode emits {"annotation": ..., "answer": ...} with annotation matching the scalar bbox schema.