Closure / Open source

Keep the reasoning connected.

When an input changes, which analyses need to run again? Does the design still meet its requirements? Closure connects your inputs, analysis scripts, and results so you can answer both questions.

For engineers working across tools and disciplines, with project files that people and AI can read.

  1. 01 / Define

    Set your inputs

    Record your assumptions and the requirements your design needs to meet.

  2. 02 / Verify

    Run the affected work

    Change a value and rerun the analyses that depend on it.

  3. 03 / Inspect

    Understand the result

    See which requirements pass, which fail, and the evidence behind each result.

Keep using your engineering tools. Make the connections between them explicit.

In your workflow

Bring the analyses
you already trust.

Start with an existing script and the requirement it supports. Use the CLI to run it, the viewer to inspect it, or the specification to connect your own tools.

01

Connect your tools

The Closure specification describes how to record requirements, share inputs and outputs, and save evidence. Use it to make your analysis tools work together across languages and disciplines.

Read the specification ↗
02

Verify from the command line

Check a project, see what needs to rerun, and follow each analysis as it executes. The CLI and Python backend preserve the results so you can inspect them later or use them in automation.

Start with the CLI ↗
03

See how your project connects

Open the viewer to follow dependencies, edit a value, and press Verify. Trace a failed requirement back to the work behind it, or revisit an earlier result.

Explore the viewer ↗

Early release candidate

Try it on work
you already do.

Take a power budget, a capacity estimate, or another analysis you use today. Connect its inputs and outputs to a requirement, change an assumption, and see how the result changes.

We’re looking for feedback from engineers trying Closure on their own projects. Tell us where you got stuck, what took too much effort, and what you needed to understand the result.

  • Use the CLI and viewer locally with Python 3.10 or 3.12 on Windows or Linux.
  • Start with examples for power budgets, service capacity, mission design, and mechanical systems.
  • Read, adapt, and build on the specification and code under Apache-2.0.