How to hold office hours with Doctor Shi Fa Wu in your AI

Most family health worries start the same way: someone feels off, a search engine offers ten confident answers, and nobody writes down what to actually ask a doctor. Heropedia's Doctor Shi Fa Wu persona is built to interrupt that pattern. It teaches the nutrition-and-lifestyle logic a biochemistry-trained lecturer uses, then sends you to your doctor with a short, specific list instead of a guess.

This is an AI persona based on Shi Fa Wu's own book, blog posts and lecture material, with every fact and quote tied to a source on the hero page. It is not Doctor Shi Fa Wu, and neither he nor any organization he is affiliated with has any connection to or endorsement of Heropedia.

The one rule that shapes every answer

The persona never diagnoses what is wrong with you, never prescribes or names a supplement for you to take, and never tells you to start, change or stop a medication. If you ask it what your fatigue means or whether to stop a prescription, it redirects you toward the question and the test to bring to your own doctor instead.

That boundary matters more here than it would for most heroes. Shi Fa Wu's own writing argues, in his words, that "no medication can possibly cure chronic disease" — a personal position, not medical consensus, and the hero page says so directly. His publisher also lists him as technical supervisor at a company that brought the Gerson regimen, a diet-based alternative cancer therapy, to China. The persona does not repeat his claims of cured tumors as fact: the hero page cites the U.S. National Cancer Institute's own conclusion that "the data that are available are not sufficient to warrant claims that the Gerson therapy is effective as an adjuvant to other cancer therapies or as a cure." If a session drifts toward cancer or any serious diagnosis, the persona's job is to hand you questions for an oncologist, not a regimen.

Why the questions are about numbers, not feelings

His stated iron law, kept on the hero page: "Disease is not an enemy, but a warning signal from the body." His other stated tenet: "Most chronic diseases come from long-term wrong nutritional supply and lifestyle." Whatever you think of those positions, they explain why the persona keeps pushing past "I feel tired" or "we eat pretty healthy" toward a number: a fasting glucose reading, a sodium tally, an hour you woke up. Vague feelings cannot be taken to a doctor and acted on; a written observation can.

Three ways to start

ChatGPT, Gemini or any chat window. Open the Doctor Shi Fa Wu page, press Copy the Shi Fa Wu office-hour prompt, paste it into a new chat and send. The persona is inside the prompt, so there is nothing to install.
Claude. Add Heropedia once as a custom connector with the URL https://www.heropedia.org/mcp (Settings, then Connectors). After that, say "Hold office hours with Doctor Shi Fa Wu on what to ask my doctor" in any chat. The setup guide walks through each step.
Claude Code, Codex or Gemini CLI. Install the office-hour skill with the command on the hero page, then say "Office hour with Doctor Shi Fa Wu before my next checkup." Your notes stay on your machine.

The six questions it will ask

Bring one real concern — yours or a family member's — and have rough details ready, because the persona pushes until an answer is a specific, reportable observation.

What is the specific, measurable symptom or test result that brought you here, and what question about it will you actually bring to your own doctor this week? It wants one concrete number or finding, not "I feel tired."
Before you touch any medication, what will you ask your doctor instead of deciding on your own to start, stop or change a dose? A question aimed at a named prescriber, not a plan to act alone.
What does your family's actual sodium, sugar and processed-food intake look like this week — have you written it down to show a professional? A real tally, not "we're careful."
Which specific nutrient or deficiency are you worried about, and what test will you ask your doctor to run before buying any supplement? A named nutrient and a named test, not "something for immunity."
What are your family's actual sleep and wake times, and what pattern have you noticed — how will you describe that to a doctor rather than to me? Hours and a pattern, not "our sleep is bad."
If you're drawn to a natural or alternative approach instead of, or alongside, a treatment your doctor recommended, what two questions will you ask about risks and interactions before changing anything? Two concrete questions for a named doctor, not a decision made in this chat.

How to get a useful session

Bring a real number, not a feeling. The persona is written to treat "I don't feel great" as the start of the conversation, not an answer it can work with.
Write down the questions, not a verdict. The session ends with a short list for your next appointment, never a diagnosis or a product to buy.
Keep any current treatment in view. If you're under a doctor's care, the persona's own rule is to ask about interactions before changing anything, never to talk you into a change here.
Treat "natural" claims as something to verify, not something to trust. The hero page itself pairs his advocacy with the mainstream medical pushback on it — bring both to your doctor.

Try another hero on the same question

A different angle on the same kind of conversation comes from Ni Haixia, whose persona teaches classical Chinese medicine's diagnostic logic under the same rule: observations for you to report, never a diagnosis from the chat. Every hero page offers the same three ways to start, so switching personas costs nothing but a new chat.