Skip to content

GG: Configurable Workflow Engine #39

@srosenbauer

Description

@srosenbauer

GG: Configurable Workflow Engine

Problem

GG’s opinionated development workflow provides a clear, structured way to build complex features with AI, and many teams benefit from that consistency out of the box. However, some teams want the ability to tailor the workflow to better fit their existing processes and requirements.

Today, GG operates as a fixed system with:

  • A hardcoded 6-step workflow (scaffold → discuss → research → plan → execute → verify)
  • Discussion questions defined by the Joggr team across predefined probe areas
  • A fixed validation pipeline
  • A defined project structure and file conventions

Teams that want to adapt GG to their own workflows currently don’t have a way to do so within the product.


What we're releasing

GG Configurable Workflow Engine introduces a user-facing configuration layer that allows teams to customize how GG operates.

This will be powered by an extended config system (e.g., gg.config.yaml or enhancements to settings.json) and will support:

  • Workflow customization: Reorder, skip, or add workflow steps, and define custom workflows beyond the default sequence
  • Discussion customization: Define custom question sets per probe area and tailor prompts to team-specific needs
  • Validation pipeline configuration: Customize what checks run during execute/verify, including team-specific validations
  • Template overrides: Customize the structure and content of generated .gg workflow artifacts (discuss, research, plan, phase docs)

This builds on the existing configuration system (deep merge via settings.json) and extends it to support workflow-level customization, making GG adaptable to any team’s development process.


Expected outcome

Teams can choose how opinionated or flexible they want GG to be.

  • Developers can keep the default workflow or tailor it to match their existing SDLC
  • Organizations can enforce team-specific requirements directly in the workflow
  • AI-driven development becomes more flexible while preserving the benefits of structure
  • GG evolves from a single prescribed workflow into a configurable platform that supports a wider range of teams and use cases

Ultimately, this allows GG to serve both teams who want strong defaults and those who need deeper customization.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    Status

    Upcoming

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions