Aura

Aura is a self-diagnostic app for pain that uses an adaptive questioning engine inspired by Akinator to narrow down plausible medical conditions based on user input. The description states it began as a solo project by Peik Gabriel, who was motivated by personal experience with chronic pain after an assault. It features an anatomical map for pain location, structured reasoning of candidate conditions, and a Care Card output. The system is built with Next.js, React, TypeScript, and integrates GPT-5.6 for natural language interpretation.

The author claims the tool avoids traditional chatbot approaches by focusing on reasoning rather than generating answers post-hoc. It uses a candidate graph that updates dynamically based on user responses, aiming to produce an evidence-traceable diagnostic path. However, there is no evidence of revenue, customers, or adoption beyond the single developer's account.

The single most important open question is: What is the actual clinical validity and safety of the reasoning engine?

The description states that Aura is a medical self-diagnostic app for identifying pain. It allows users to point to where it hurts, answer adaptive questions, and receive a structured, evidence-traceable starting point to bring into care.

It uses an anatomical map for pain location and an adaptive questioning engine that selects the next question based on which would separate remaining candidates most effectively.

The system builds upon a candidate graph containing plausible pain generators, clinical questions, possible responses, and rules for selecting the next useful question. It is described as having a deterministic diagnostic core with an intelligent, human-facing layer around it.

It generates a Care Card summarizing the pain location, relevant context, important responses, strongest pattern, and care timeframe.

Evidence: The author's own write-up.

Inference: The product is built as a web application using Next.js, React, TypeScript, and integrates GPT-5.6 for natural language interpretation.

The description states that Aura began with the author’s personal experience of chronic pain following an assault. It was inspired by Akinator, a guessing game, but applied to medical self-diagnosis.

The author claims that Aura is not a chatbot that generates convincing answers post-hoc, but instead builds the reasoning itself. It aims to reduce the burden on clinicians by handling cataloguing work so they can focus on human interaction.

It positions itself as a tool that turns “something hurts” into a structured, traceable story, allowing users to better communicate their condition before seeing a doctor.

Evidence: The author's own write-up.

Inference: The positioning evolved from personal need to a broader vision of improving clinical workflows through technology.

The description states that Aura is intended for individuals experiencing pain who want to better understand what might be causing it and how to communicate that to healthcare providers.

It targets people who have had unsatisfactory medical appointments, where doctors failed to provide clear explanations or solutions. The system is designed to help users prepare for consultations by giving them a clearer understanding of their condition.

Evidence: The author's own write-up.

Inference: The target customer is likely someone with chronic or unexplained pain who seeks clarity and better communication with clinicians.

Not evidenced.

The description does not mention any pricing model, monetization strategy, or business model. There is no indication of whether the app will be free, paid, or offered through partnerships.

Evidence: The author's own write-up.

Inference: No commercial structure is described beyond the solo development effort.

The system is built with Next.js 16, React 19, TypeScript, and Zod. It uses Firebase for hosting and integrates Codex as an engineering collaborator during development.

It includes a visual experience using Canvas, and utilizes GPT-5.6 server-side assistant to interpret natural language or explain results conversationally.

The reasoning engine is described as deterministic, inspectable, and capable of handling supporting, conflicting, unsure, and unknown evidence states.

Assessment data remains in browser memory with no account or personal health-history database required.

Evidence: The author's own write-up.

Inference: The technical stack suggests a modern web application with AI integration and structured reasoning capabilities.

Not evidenced.

There is no mention of users, customers, revenue, or adoption metrics. The project is described as a solo effort by one developer, submitted to a hackathon, without any indication of traction or market validation.

Evidence: The author's own write-up.

Inference: No signs of product-market fit or commercial traction are evident.

Not evidenced.

The description does not reference existing tools or competitors in the medical self-diagnosis space. It only mentions inspiration from Akinator, without comparing to other diagnostic apps or platforms.

Evidence: The author's own write-up.

Inference: No competitive landscape is described beyond the inspiration source.

1. Clinical Validity and Safety: The description does not provide evidence of clinical validation or safety testing. The reasoning engine may be unvalidated in real-world use, raising concerns about misdiagnosis or harm.
2. Single Developer Limitation: The product is described as a solo effort by one developer, which raises questions about scalability, long-term maintenance, and feature completeness.
3. Lack of User Testing or Feedback: There is no evidence of user testing, feedback loops, or iterative improvements beyond the author’s personal experience.
4. No Data Persistence or Integration: The system stores data only in browser memory and does not integrate with existing health systems or databases, limiting its utility for ongoing care.
5. Unverified Claims: All claims about functionality, reasoning, and outcomes are self-reported and unverified.

Evidence: The author's own write-up.

Inference: These risks stem from the lack of independent verification, clinical oversight, and real-world usage data.

1. What is the clinical basis for the candidate graphs used in the diagnostic engine?
2. Has the reasoning engine been tested with real users or validated by medical professionals?
3. How does the system handle edge cases or ambiguous inputs from users?
4. Are there plans to integrate with electronic health records or other healthcare systems?
5. What are the privacy and data handling policies for user information?
6. How is the accuracy of the candidate ranking and explanations ensured over time?

Evidence: The author's own write-up.

Inference: These questions aim to probe the unverified claims and validate assumptions made in the description.