One always-on agent per patient, not one app per problem — Apr 23, 2026
Patient scribe inside a patient portal
Max showed a patient scribe: the patient enters or scans data that arrived on paper, it lands as unverified, and a practitioner confirms it later — which is the whole point, since the case he's building for is records that reach a patient in anything but FHIR.
Everything is validated against profiles by his TypeScript library that converts StructureDefinitions, and the portal handles DICOM upload with a viewer, document scanning with extraction, and time-limited read-only sharing links. He flagged consent as a bigger topic than a share link and worth its own session.
Health Claw — always-on personal health agent
Eugene's Health Claw is a set of skills and agent templates that put a patient's own data behind an always-on agent they control. The agents are deliberately segmented by domain — the one managing prescriptions is not the one giving fitness advice — and data from SMART apps, wearables, claims, labs and pharmacies passes through a guardrail layer that strips PHI first, so a compromised instance doesn't hand over the whole health record.
A curator agent checks data quality and tells the patient when a code is wrong or outdated rather than letting it quietly inform medical advice.
The story carrying the thesis — a surprise $1,000 bill
Gene had the agent pull his family's claims history, analyse the charges against the health plan's own authorisation guidelines and produce a specific action plan — talk to the provider, talk to the insurer — and it saved the whole amount, which he was quick to say is nothing next to what people go bankrupt over.
The burden has been shifted onto patients. A seriously ill patient needs a graduate-level understanding of the healthcare system to navigate her own care. An agent aligned to the patient is a way to shift some of it back.
The argument against apps and third-party wallets
No patient is going to install an app per problem or wire up MCP servers themselves — they aren't aware enough to, and won't be. What works is one always-on agent reachable over Telegram or WhatsApp that hooks into the growing MCP ecosystem on their behalf.
Gene thinks that makes referral management — currently a week of repeated phone calls — the next thing to fall.
He also made the ownership case plainly: HIPAA protects data on the provider's side, so once you download it and hand it to a third-party app, that company can do as it likes. Self-hosting is the only real control.
The recurring spec argument returns
Alex Szilagyi noted BMAD's persona-based roles looked like Gene's agent structure. Gene's approach working solo is to skip heavy upfront specs — you go through that thinking anyway while planning — and instead run /init every session so the project's CLAUDE.md compounds, working in small chunks because the more context you hand over at once the harder it is to hit the target.
The living spec is exactly that memory, but shared. It lets me run several repo clones as independent developers on parallel features — and it matters more, not less, when you move between model providers.