Importing Node Models

Node models power the workspace. The backend collects node models from two sources and returns a merged list to the frontend.

Sources

1. AutoAPMS Node Manifests: discovered via auto_apms_behavior_tree_core resources when AutoAPMS is installed. For this, the backend uses get_node_manifest_resource_identities() from auto_apms_behavior_tree_core.resources.
2. BehaviorTree.CPP Native Nodes: discovered via the native registry exposed by BehaviorTree.CPP helper packages. AutoAPMS Studio utilizes get_native_node_model() from auto_apms_behavior_tree.resources to access the native node registry.

How it works in detail

• AutoAPMSLoader reads available manifest identities using get_node_manifest_resource_identities() and converts NodeManifestResource data into the backend NodeModel format.
• NativeTreeLoader calls get_native_node_model() and converts native node descriptions to NodeModel entries.
• The nodes router (GET /api/v1/node_modules/) concatenates both lists and returns a NodeModelList.

Development Notes

• NodeModel contains port information (input_ports, output_ports, input_output_ports), defaults, aliases and other metadata used by the frontend to render node configuration forms.
• To add another source, implement an adapter that returns list[NodeModel] and include it in the router aggregation.

Caching and Standalone Node Editor

• The standalone web editor uses cached node models or built-in defaults if no backend is available. The backend is the source of truth when present.
• You can export your currently loaded node models to a JSON file and import them into the standalone editor. This can be done through the settings menu.

Code Locations for Development

• AutoAPMS Adapter: auto_apms_studio/backend/app/adapters/autoapms.py
• Native Node Model Adapter: auto_apms_studio/backend/app/adapters/cpp_nodes.py
• Router: auto_apms_studio/backend/app/routers/nodes.py
• Models: auto_apms_studio/backend/app/models/nodes.py