perceptic-workflow-sdk 0.50.98

Python SDK for Perceptic Workflow definitions - cross-language Temporal workflow types

Perceptic Workflow SDK

Python SDK for Perceptic Workflow definitions. This package provides cross-language Temporal workflow types that are compatible with the Java workflow definitions in perceptic-core-client.

Installation

uv add perceptic-workflow-sdk

Usage

Runtime Input Contract

Workflows started through Perceptic Core receive a typed envelope:

• AgentInput[T] where:
  • run_rid identifies the run
  • user_id identifies the authenticated caller
  • payload contains your workflow-specific input model T

On the wire, Perceptic Core sends camelCase (runRid, userId, payload). In Python, use snake_case (run_rid, user_id, payload) per PEP 8. The SDK handles alias mapping automatically.

Workflow Events

from perceptic_workflow import (
    WorkflowEvent,
    InfoEvent,
    CheckpointEvent,
    UserInputEvent,
    UserInputRequestEvent,
    WorkflowCheckpointStatus,
)

# Create events
info_event = InfoEvent(
    event_id=1,
    type="progress",
    payload={"step": "processing"},
    timestamp=datetime.now()
)

Implementing Workflows

The SDK provides a mixin class for implementing Perceptic-compatible workflows:

from pydantic import BaseModel
from temporalio import workflow
from perceptic_workflow import AgentInput, PercepticWorkflow, WorkflowEvent


class ResearchInput(BaseModel):
    query: str
    context: str

@workflow.defn
class MyWorkflow(PercepticWorkflow):
    def __init__(self):
        self._events: list[WorkflowEvent] = []
        self._paused = False

    @workflow.run
    async def run(self, input: AgentInput[ResearchInput]) -> dict:
        # Envelope fields use snake_case in Python.
        run_rid = input.run_rid
        user_id = input.user_id
        request = input.payload

        if request is None:
            raise ValueError("payload is required")

        return {
            "runRid": run_rid,
            "userId": user_id,
            "query": request.query,
            "context": request.context,
        }

    @workflow.update(name=PercepticWorkflow.UPDATE_SUBMIT_USER_INPUT)
    async def submit_user_input(self, inputs: dict) -> None:
        self._paused = False
        # Handle user input

    @workflow.update(name=PercepticWorkflow.UPDATE_INTERRUPT)
    async def interrupt(self, reason: str) -> None:
        # Handle interruption
        pass

    @workflow.query(name=PercepticWorkflow.QUERY_IS_PAUSED)
    def is_paused(self) -> bool:
        return self._paused

    @workflow.query(name=PercepticWorkflow.QUERY_GET_EVENTS)
    def get_events(self, after_event_id: int | None = None) -> list[WorkflowEvent]:
        if after_event_id is None:
            return self._events
        return [e for e in self._events if e.event_id > after_event_id]

Decorator Validation

PercepticWorkflow no longer performs runtime validation of Temporal handler decorators during class creation. This avoids import-time failures for abstract workflow bases and mixin-based inheritance patterns.

Handler contract enforcement is expected to happen through linting and type-check tooling in your development pipeline:

uv run ty check
uv run workflow-typecheck

Troubleshooting

• ValidationError: Field required ... run_rid/user_id during workflow activation usually means your @workflow.run signature expects the wrong shape.
• Use AgentInput[YourPayloadModel] for the first workflow argument, not a flattened domain model.
• For run-start requests, send runRid and inputs; userId is derived from auth context by Perceptic Core.

Compatibility

This package is designed to be compatible with:

• perceptic-core-client (Java)
• perceptic-core-workflow-runtime (Java)

The workflow definitions and event types are generated from the same JSON Schema sources to ensure cross-language compatibility.