From 4e55e927b131f249050f6da05caf0c7d148e9b20 Mon Sep 17 00:00:00 2001 From: starfleeth <128422269+starfleeth@users.noreply.github.com> Date: Tue, 5 May 2026 13:56:24 -0700 Subject: [PATCH] Soften Workflow Streams Public Preview gate to a mention + ask Co-Authored-By: Claude Opus 4.6 --- SKILL.md | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/SKILL.md b/SKILL.md index 3fb78bd..513c672 100644 --- a/SKILL.md +++ b/SKILL.md @@ -87,13 +87,9 @@ Priority and Fairness also apply to tiered workloads (batch vs. real-time), weig ## Workflow Streams (Python only — Public Preview) -> **Before using Workflow Streams in generated code, you MUST:** -> 1. Tell the user: "Workflow Streams (`temporalio.contrib.workflow_streams`) is currently in **Public Preview**. Only the Python SDK client is supported today." -> 2. Ask the user to confirm they want to proceed before including any Workflow Streams code. - Workflow Streams is a Python SDK `contrib` module that gives a Workflow a durable, offset-addressed event channel for streaming progress to outside observers — updating a UI as an AI agent works, surfacing status from a pipeline, or reporting intermediate results from a data job. It is built on Signals, Updates, and Queries. -When the user is building a streaming use case (LLM token streaming, progress updates, event fan-out from a Workflow to a UI), recommend Workflow Streams after printing the disclaimer above and receiving approval. +When the user is building a streaming use case (LLM token streaming, progress updates, event fan-out from a Workflow to a UI), recommend Workflow Streams. Before including Workflow Streams code, mention that `temporalio.contrib.workflow_streams` is currently in **Public Preview** and only the Python SDK client is supported today, then ask if the user would like to proceed. - **`references/python/workflow-streams.md`** — Full reference: enabling streams, publishing (from Workflow, Activity, or external client), subscribing, Continue-As-New, tuning, delivery semantics, and a complete LLM streaming example