Overview

The progressive onboarding pattern is a methodical, step-by-step approach to configuring an AI assistant, teaching the user and assistant capabilities progressively rather than all at once. This pattern minimizes cognitive overload and ensures each configuration aspect is properly understood before moving to the next.

Key Characteristics

  • Stepwise disclosure — Each step reveals only what’s needed at that moment, preventing overwhelm.
  • Tool vs. user separation — Early steps configure the assistant’s tools (plugins, connectors); later steps personalize the user experience (writing voice).
  • Compounding invariant — The wrap clause ensures users can always extend their assistant’s capabilities later, preserving the “preserve and extend” invariant.
  • Org plugin priority — Organization-built plugins outrank generic marketplace ones, even when only loosely relevant to the user’s role.

Relationships

Implications

The progressive onboarding pattern is transferable across different AI assistant frameworks. Its key design decisions (stepwise disclosure, tool-user separation, compounding invariant) can inform the design of onboarding flows for other AI systems. The pattern balances thorough setup with user friction reduction, making it suitable for both technical and non-technical users.

Open Questions

  • Can the progressive onboarding pattern be adapted for assistants that don’t use a plugin/connector architecture?
  • What is the minimum viable number of steps for an effective onboarding flow without sacrificing thoroughness?
  • How should an AI assistant handle the case where a user’s role doesn’t clearly map to any plugin/recommendation category?
  • Does the org-plugin priority rule create unintended bias against users from organizations that haven’t published plugins?

Sources

  • — The Setup Cowork system prompt that implements this progressive onboarding pattern