A portable, de-identified chronology built from one synthetic case file — drop it into any AI assistant and start asking questions, citations included.
The source record RITA read, and the portable pod she built from it — grab one or both.
The raw record RITA read — every report and the RITAPod below were built from this single file.
Download Case File (PDF) →The de-identified chronology, plus everything you need to run it — guides, a local runner, and the launcher script.
Download RITAPod (.zip) →The zip unpacks into one main file — DropMeInYourFavoriteAI.json — along with a printed guide and an optional local runner. The grounding and citation rules travel inside the file, so most of the time there's nothing else to paste.
python ritapod_runner.py from the folder. That path needs Python 3.10+ installed and works great with a local model via Ollama.Four kinds of questions the pod handles well — from a quick overview to strategy. We ran each of these against the actual pod before publishing them here.
This is the Neutral-lens RITAPod for the Sutton demo case — Citation Contract grounding only, no plaintiff or defense emphasis. The lens sets default summary emphasis, not a cage: ask a direct question and the pod will answer it either way, citations included.
Patient identifiers — name, DOB, MRN, SSN, phone, email — are scrubbed and machine-verified before packaging; export is blocked if residual patient PHI is detected. Provider and facility names are intentionally retained for litigation analysis. As with any case data, your firm remains responsible for HIPAA compliance and any vendor BAAs when it's used with a third-party AI service.
This is a synthetic demo case built to be shown, not a real patient record.