Your next analysis.
A smaller first step.

Start from a working Python example. Change the science, keep the contracts, and use the same path whether you write the code yourself or work with an agent.

Four commands.
One real result.

Run these in a local project checkout with Python 3.11+ and uv installed. The generated example contains synthetic data and needs no cloud account.

  1. Start with a templateA function, a manifest, and a sample trace.
  2. Edit and validateExplicit inputs, units, versions, and outputs.
  3. Run your trusted codeWrite a new result file with provenance.
  4. Preview and reviewReview the saved output; share a PR with tests.
Local terminalLocal SDK / 0.1.0
uv sync --extra dev
uv run hb init my-analysis
uv run hb validate my-analysis
uv run hb run my-analysis --input my-analysis/sample.json --output result.json
my-analysis/
  main.py          ← your function
  extension.json   ← typed contract
  sample.json      ← synthetic example
  README.md        ← local workflow

result.json, written by the last command (abridged)

{
  "output": {
    "joint": "knee_flexion",
    "side": "left",
    "metrics": [
      { "key": "minimum", "value": 10.0, "unit": "deg" },
      { "key": "maximum", "value": 60.0, "unit": "deg" },
      { "key": "range_of_motion", "value": 50.0, "unit": "deg" },
      { "key": "sample_mean", "value": 34.79338842975206, "unit": "deg" }
    ],
    "chart": {
      "kind": "line", "x_unit": "s", "y_unit": "deg",
      "points": [ ...121 samples ]
    }
  },
  "provenance": {
    "module": "example-lab/knee-summary",
    "module_version": "0.1.0",
    "sdk_version": "0.1.0",
    "execution": "local-trusted",
    "input_sha256": "aa3042ad...",
    "source_sha256": "cedf03e3..."
  }
}

Codex or Claude Code.
The same foundation.

Shared instructions and deterministic CLI commands are available now. A packaged contribution skill, MCP server, and one-click plugin are planned.

Create a knee-angle summary module from the project template. Use synthetic data, add known-answer and invalid-input tests, validate the output, and prepare a reviewable change with a changelog entry.
Local execution runs with your user permissions.

Only run Python code you trust. The local runner has a timeout and contract checks, but it is not a security sandbox. Do not use patient data or production credentials in this preview.

Accessibility

Make every MotionCommons page comfortable for you. Settings apply on this browser.

Text size

Your training, at your pace.

Reading & movement
Audio support

Spoken audio is unavailable in this browser. All instructions remain available as text.

Read this page speaks up to 1,800 characters of visible page text. Embedded dashboards have their own Read dashboard button.