fix(dojo): re-enable agentic generative UI test

Signed-off-by: Tyler Slaton <[email protected]>

Author: tylerslaton

typescript-sdk/apps/dojo/e2e/tests/pydanticAITests/agenticGenUI.spec.ts

@@ -3,7 +3,7 @@ import { AgenticGenUIPage } from "../../pages/pydanticAIPages/AgenticUIGenPage";
 
 test.describe("Agent Generative UI Feature", () => {
   // Flaky. Sometimes the steps render but never process.
-  test.fixme("[PydanticAI] should interact with the chat to get a planner on prompt", async ({
+  test("[PydanticAI] should interact with the chat to get a planner on prompt", async ({
     page,
   }) => {
     const genUIAgent = new AgenticGenUIPage(page);

typescript-sdk/integrations/pydantic-ai/examples/server/api/agentic_generative_ui.py

@@ -58,6 +58,7 @@ class JSONPatchOp(BaseModel):
         - Do NOT repeat the plan or summarise it in a message
         - Do NOT confirm the creation or updates in a message
         - Do NOT ask the user for additional information or next steps
+        - Do NOT leave a plan hanging, always complete the plan via `update_plan_step` if one is ongoing.
 
         Only one plan can be active at a time, so do not call the `create_plan` tool
         again until all the steps in current plan are completed.