Explains specialized Synapse action classes for specific workflows. Use when the user mentions "BaseTrainAction", "BaseExportAction", "BaseUploadAction", "BaseInferenceAction", "BaseDeploymentAction", "AddTaskDataAction", "train action", "export action", "upload action", "inference action", "deployment action", "pre-annotation", "add_task_data", "autolog", "get_dataset", "create_model", or needs workflow-specific action development help.
Explains specialized Synapse action classes for ML workflows like training, inference, and deployment.
/plugin marketplace add datamaker-kr/synapse-claude-marketplace/plugin install synapse-plugin-helper@synapse-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
references/add-task-data-action.mdreferences/export-action.mdreferences/inference-action.mdreferences/progress-categories.mdreferences/train-action.mdreferences/upload-action.mdSynapse SDK provides specialized base classes for common ML workflows. Each extends BaseAction with workflow-specific helper methods and default settings.
| Class | Category | Purpose |
|---|---|---|
BaseTrainAction | NEURAL_NET | Training models |
BaseExportAction | EXPORT | Exporting data |
BaseUploadAction | UPLOAD | Uploading files |
BaseInferenceAction | NEURAL_NET | Running inference |
BaseDeploymentAction | - | Ray Serve deployment |
AddTaskDataAction | PRE_ANNOTATION | Pre-annotation workflows |
# Training - autolog, get_dataset, create_model
class TrainAction(BaseTrainAction[TrainParams]):
def execute(self) -> dict:
self.autolog('ultralytics') # Auto-log metrics
dataset = self.get_dataset()
# ... train ...
return self.create_model('./model.pt')
# Export - get_filtered_results
class ExportAction(BaseExportAction[ExportParams]):
def get_filtered_results(self, filters: dict) -> tuple[Any, int]:
return self.client.get_assignments(filters)
# Upload - step-based workflow required
class UploadAction(BaseUploadAction[UploadParams]):
def setup_steps(self, registry: StepRegistry[UploadContext]) -> None:
registry.register(InitStep())
registry.register(UploadFilesStep())
# Inference - download_model, load_model, infer
class InferAction(BaseInferenceAction[InferParams]):
def execute(self) -> dict:
model = self.load_model(self.params.model_id)
return {'predictions': self.infer(model, self.params.inputs)}
# Pre-annotation - convert_data_from_file, convert_data_from_inference
class PreAnnotateAction(AddTaskDataAction):
def convert_data_from_file(self, primary_url, ...) -> dict:
return {'annotations': [...]}
All specialized actions (except Deployment) support two modes:
execute() for straightforward workflowssetup_steps() for complex multi-step workflows with rollback# Simple mode
class SimpleTrainAction(BaseTrainAction[Params]):
def execute(self) -> dict:
return {'weights_path': '/model.pt'}
# Step-based mode
class StepTrainAction(BaseTrainAction[Params]):
def setup_steps(self, registry: StepRegistry[TrainContext]) -> None:
registry.register(LoadDatasetStep())
registry.register(TrainStep())
registry.register(UploadModelStep())
Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
Search, retrieve, and install Agent Skills from the prompts.chat registry using MCP tools. Use when the user asks to find skills, browse skill catalogs, install a skill for Claude, or extend Claude's capabilities with reusable AI agent components.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.