PatternsSDK API

Plan

Loading pattern preview

Problems solved

Technical challenges this implementation handles out of the box.
  • Stream AI-generated execution plans into a collapsible plan UI
  • Convert plan data parts back into model-visible text for edits
  • Generate structured plan updates through a dedicated tool
  • Show shimmer loading while plan steps arrive

Use cases

Products and workflows this pattern is designed to support.
  • Project planning copilots
  • Marketing campaign plan builders
  • Technical migration checklists
  • Task breakdown assistants

Setup

Requirements, wiring steps, and what this pattern adds to your project.

Getting started

Pick how you want to pull this pattern in. Then wire env vars and routes the same way.

  1. 1

    Install from the preview toolbar

    Shadcn Icon

    Copy the install command above and run it in your project — Pro patterns include a short-lived token.

  2. 2

    Add environment variables to .env.local

    OPENAI_API_KEY

  3. 3

    Point client fetches at your API routes

    Replace preview `/view/.../api/...` paths with `/api/...` in the installed files.

  4. 4

    Customize the agent and tool files

    Adapt prompts, tools, and stop conditions for your product — Copy for AI in the toolbar helps seed that work.

  5. 5

    Run your dev server and open the pattern route

    Install dependencies if needed, then start the app and verify responses.

Environment variables

Capabilities

AI SDK APIs

ToolLoopAgentcreateUIMessageStreamcreateUIMessageStreamResponseconvertToModelMessagesvalidateUIMessagesstreamTexttoolgateway

Providers

OpenAI

Files added

7 files
  • app/page.tsx
  • app/layout.tsx
  • components/plan-demo.tsx
  • lib/types.ts
  • lib/plan-agent.ts
  • lib/tools/plan-generator.ts
  • app/api/plan/route.ts

Dependencies

8 total

npm packages

4
AISDKlucide-reactNext.jsReact

Shadcn IconRegistry components

4
cardcollapsiblebuttonempty

Critical files

  • plan/route.ts

    API route for the plan agent. Uses createUIMessageStream with custom PlanUIMessage type. Implements convertDataPart to transform plan data parts back into text so the model can see and edit existing plans. Extracts generatePlan tool output on finish to write structured data-plan-info parts.

  • lib/plan-agent.ts

    Defines the ToolLoopAgent for plan generation. Configures 5 tools (generatePlan, executeStep, updatePlan, reviewPlan, buildPlan) with detailed workflow instructions. Uses stepCountIs(20) stop condition.

  • tools/plan-generator.ts

    Implements 5 generator-based plan tools using async *execute pattern for progressive streaming. Each tool uses nested streamText calls for AI-powered content generation.

Requirements

Complexity
Beginner
Setup time
~10 minutes
Node
24+
Memory
512MB