Create a new diagram with specified nodes and edges.

A diagram is the primary unit of reasoning, representing a directed acyclic graph.

Args: nodes: List of node specifications. Each must have: - id: Unique identifier (string) - label: Human-readable label (string) - type: One of 'operation', 'terminal', 'control', 'composite' - metadata: Optional dict of additional properties - embedding: Optional list of floats for semantic search

edges: List of edge specifications. Each must have:
    - source: Source node ID (string)
    - target: Target node ID (string)
    - label: Relationship label (string)
    - weight: Optional numeric weight (default: 1.0)
    - properties: Optional dict of additional properties

Returns: dict with diagram_id, num_nodes, num_edges, validity

Load and return complete diagram structure.

Args: diagram_id: ID of diagram to load

Returns: dict with full diagram structure or error

Explore diagram structure using breadth-first traversal.

Systematically visits nodes level by level, useful for understanding structure.

Args: diagram_id: ID of diagram to explore start_node: Starting node ID max_depth: Maximum exploration depth (default: 5)

Returns: dict with explored_nodes, edges_traversed, total_explored

Find guided path from start to goal node.

Uses shortest-path algorithm for optimal traversal.

Args: diagram_id: ID of diagram start_node: Starting node ID goal_node: Target node ID heuristic: 'distance' (default), 'reward', or 'semantic_similarity'

Returns: dict with path, cost, num_steps, found flag

Analyze reachability from source node.

Computes all nodes reachable from source and their distances.

Args: diagram_id: ID of diagram source: Source node ID targets: Optional specific nodes to check

Returns: dict with reachable_nodes list, distances, target_reachability

Wander the diagram based on a curiosity metric using NavigationMemory.

Prioritizes nodes that have a low exploration_count.

Args: diagram_id: ID of diagram start_node: Starting node ID steps: Number of steps to take (default: 5) temperature: Exploration factor (0.0 = greedy unvisited, 1.0 = highly random)

Returns: dict with path, steps_taken, and updated exploration_counts

Extract a subgraph into a composite node.

Args: diagram_id: Source diagram ID node_ids: List of nodes to extract composite_label: Label for the new composite node

Returns: dict containing success status, new diagram ID, and new composite node ID

Apply a formal Double-Pushout (DPO) rewrite rule to a diagram.

Args: diagram_id: Target diagram ID rule_spec: Serialize RewriteRule dict with 'lhs' and 'rhs' patterns match_mapping: Mapping of rule LHS node IDs to diagram node IDs (from pattern_match)

Returns: dict containing success status and modified diagram stats

Export a diagram's transformation history as a structured proof.

Args: diagram_id: ID of diagram to export proof for output_format: 'text' (natural language) or 'json' (structured)

Returns: dict with proof steps and metadata

Check if two diagrams are structurally equivalent (isomorphic).

Args: diagram_id_1: First diagram ID diagram_id_2: Second diagram ID

Returns: dict with equivalence status

Explore alternative diagram states reachable via rewrite rules.

Constructs a meta-graph of structurally unique diagram configurations.

Args: diagram_id: Starting diagram ID rules: List of serialized RewriteRule dicts max_depth: Maximum BFS depth max_states: Maximum number of unique states to discover

Returns: dict with metadata about discovery and unique states

Find all pattern matches within a diagram.

Implements subgraph matching to locate structural patterns.

Args: diagram_id: ID of diagram to search pattern: Pattern spec with 'nodes' (dict) and 'edges' (list) Each edge tuple: (source_id, target_id, constraints_dict)

Returns: dict with matches list, num_matches

Compute structural metrics on diagram.

Computes graph-theoretic properties useful for understanding complexity.

Args: diagram_id: ID of diagram metrics: List of metrics to compute. Options: - 'chain_length': Longest path in DAG - 'branching_factor': Average out-degree - 'density': Overall connectivity - 'num_nodes': Node count - 'num_edges': Edge count - 'is_dag': Whether diagram is acyclic

Returns: dict with computed metrics

Search for nodes semantically similar to a target embedding.

Args: diagram_id: ID of diagram target_embedding: The embedding vector (list of floats) to compare against top_k: Maximum number of results to return (default: 5) threshold: Minimum cosine similarity score to include (default: 0.5)

Returns: dict with similar nodes and similarity scores

Force an immediate sync of a specific diagram to disk.

Args: diagram_id: ID of diagram to save

Returns: dict containing success status

List all diagram IDs currently persisted on disk.

Returns: dict with list of saved diagram IDs

Delete a diagram from both active memory and persisted storage.

Args: diagram_id: ID of diagram to delete

Returns: dict containing success status

Return server capabilities and status information.

Returns: dict with version, capabilities, active_diagrams, resources