ShipShape

Stop hunting for the critical path! ShipShape is the AI co-pilot that eats your 50-page requirement docs for breakfast and builds a live mindmap of the actual work to be done.

Comment

ShipShape — About the Project

Inspiration

Starting new work often means long requirement PDFs, unclear ownership, and scattered status across tools. We set out to shorten the path from requirements to tasks to delivery, and to give teams a live, visual model of the work that stays in sync with reality.

What it does

ShipShape is a developer tools platform with three core views:

  • Team Dashboard for daily flow, assignments, and status
  • Project Onboarding to set up storage and guide new contributors
  • Project Mindmap to visualize structure, dependencies, and progress

AI support reads requirement documents, proposes tasks, and explains assignments, while keeping a human in the loop. The system persists state so views remain consistent as work evolves.

How we built it

Frontend

  • Team Dashboard built with Tailwind CSS and lucide icons
  • Project Onboarding that initializes storage and loads projects through the API
  • Project Mindmap with drag and drop nodes, connections, auto layout, and REST sync for read and write

Backend

  • Python 3.11 with Flask 3.x and Flask SQLAlchemy
  • Stateless JSON endpoints for projects, nodes, connections, and assignments
  • SQL database behind SQLAlchemy for reliable CRUD

Where we used Claude

  • Requirements analysis and task breakdown from PDFs, loaded through the API and surfaced in the Dashboard and Mindmap
  • Assignment reasoning that generates short rationales based on skills, workload, and priority
  • Code quality aids for documentation stubs and test templates attached to tasks
  • Mindmap seeding from generated tasks and relationships

Challenges we ran into

  • Keeping the Mindmap consistent during fast edits, drags, and connection changes
  • Producing tidy layouts while minimizing edge crossings and handling smooth transitions after reflow
  • Evolving the schema while still supporting assignment notes and audit trails
  • Telling an end to end story in under ten minutes without overpromising by clearly separating implemented features from AI components

Accomplishments that we are proud of

  • A reliable core that stands on its own with AI layered in cleanly
  • A live Mindmap that persists edits through REST and stays in sync across views
  • Transparent AI rationales shown next to assignments
  • A small, stable API surface that made iteration fast and safe

What we learned

  • Keep the AI boundary thin and explicit so the product is dependable even when AI is offline
  • Design for synchronization first when multiple views edit the same data
  • Short, structured onboarding reduces time to first contribution
  • A minimal schema plus audit trails keeps the system flexible without losing context

What is next for ShipShape

  • Conflict handling and offline tolerant sync for Mindmap edits
  • Template libraries for common project types and requirement patterns
  • Deeper assignment signals from historical throughput and skill matrices
  • Export and import bridges to Jira and GitHub Issues
  • Stronger testing around layout stability and large graphs
  • Enterprise needs such as SSO and role based access control

Built With

Share this project:

Updates