Informed Patient skill
A Claude skill for health evidence reviews: guided interview, evidence red flags and a ranked question list for your appointment
- 191 ★ on the repository
- Last commit
- CC-BY-4.0
Extended access
Why this level
- A full review means describing symptoms and medical history, and that information goes to an AI service.
- An AI can cite a source that does not exist or overstate the strength of evidence, so the output is not a medical recommendation.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/informed-patientDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- A full review means describing symptoms and medical history, and that information goes to an AI service.
- An AI can cite a source that does not exist or overstate the strength of evidence, so the output is not a medical recommendation.
This is third-party code. Review the repository files before installing.
What it does
Informed Patient builds an evidence review around a single health question. Claude first runs a semi-structured interview: symptoms, timeline, effect on daily life, what has been tried and which diagnoses have come up. It then searches the literature along a source hierarchy: systematic reviews, clinical guidelines, primary studies, regulator data and patient-advocacy material. By default the search follows that hierarchy step by step. For complex cases that sit between several conditions there is an open mode, where the hierarchy is only a starting point. The result is saved as a "Health Evidence Review": a symptom inventory, search terms, a note for checking sources, an evidence snapshot, competing hypotheses with evidence for and against, and questions for the doctor in the order the user sets. The skill makes no diagnoses and recommends no treatment. It trains Claude to resist confirming the first guess and flags weak spots in the evidence base.
Who it is for. Patients and their relatives who are preparing for an appointment or making sense of a diagnosis and want to arrive with organized symptoms and questions.
Good fit when
- A first appointment with a new doctor is a week away and the symptoms need pulling into one picture.
- Tests and specialists have gone on for months and nothing is clear yet.
- A doctor named a diagnosis and you want to know how well studied it is and what treatment options exist.
- A new medicine was prescribed and you want to know what the research says about it.
- Several symptoms look unrelated and you want to trace them over time.
Not a fit when
- An acute condition: in a life-threatening situation call emergency services, the skill is not built for that.
- You expect a diagnosis or a treatment decision: the skill gives neither and does not replace a doctor.
- The topic is mental health or emotional support: the skill excludes it.
- Describing symptoms to an AI service is off the table for privacy reasons.
Example request
Use the informed-patient skill: fatigue and joint pain have lasted three months, help me prepare for a rheumatologist appointment.Limitations
The skill has not been clinically validated and works as a reasoning aid, not a medical instrument. It is worth opening the links in the sources section yourself, because an AI can produce a citation that does not exist. The search stops once the evidence is sufficient, so for well-studied conditions some sources are skipped; a full pass is requested separately. The hierarchy leans on US and UK sources, while WHO or national guidelines elsewhere may differ. Claude's web search cannot restrict a query to one site, so for full PubMed coverage you connect the PubMed connector. A chat with an AI does not carry medical confidentiality: HIPAA does not apply to it.
How to disable. In Claude Code remove the plugin through the `/plugin` menu. In Claude Desktop and on the web delete the skill from the skills list in settings.
FAQ
How do I turn the Informed Patient skill on?
Name it in your request: it does not fire on its own for health questions. For example: "Use the informed-patient skill, I am preparing for a neurologist appointment." Claude then starts asking about your symptoms.
How long does a session take?
One review takes 10 to 15 minutes. You can split the work and come back later, and several reviews in a row help refine the search.
What ends up in the final document?
A symptom inventory, the list of search terms, a note for checking sources, an evidence snapshot, competing hypotheses and questions for the doctor. The user sets their order.
Does the skill work without Claude Code?
Yes. For Claude Desktop and the web app download the `informed-patient-skill.zip` archive from the releases page and upload it through Create Skill, then Upload a Skill. Users report that the instructions also work as a prompt for other models, local ones included.
Do I have to share personal details?
No: the skill works without a name, date of birth, workplace or insurance details. Describing symptoms and circumstances is enough. On a work account an administrator may see the chat, so a personal account suits health topics better.
Discussion
No questions or reviews yet.
Related
SkillsA library of agent skills for medical research: literature search and appraisal, study design, data analysis and academic writing
Anthropic's official plugin for healthcare work: ICD-10-CM coding, prior authorization, FHIR access and PubMed search
Skills for pharma R&D: clinical trial design, ADaM datasets with admiral, Kaplan–Meier digitizing and statistical review.
Around 200 skills for Claude Code on bioinformatics, chemistry and biostatistics, from RNA-seq analysis to molecular docking
High-Stakes Analytics Lab
High-Stakes Analytics & Decision Lab
An evidence-gated analytics skill for high-stakes decisions: profiles data, picks a method the evidence supports, can answer with a data request
A skill retrieves Carrel evidence, reads source passages and images and helps agents cite original files in answers.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Informed Patient (this card) | no | 191 | |
| Medical Research Agent Skills | no | 2,018 | a skill library for researchers covering literature review and papers; Informed Patient leads a patient from a symptom interview to questions for the doctor |
| Claude for Healthcare | yes | 440 | an Anthropic plugin for clinical and insurance tasks such as ICD-10 codes and FHIR; Informed Patient prepares an ordinary person for an appointment |
README in short
The repository is packaged as a Claude Code plugin with a marketplace and duplicated as an archive for Claude Desktop. It includes an `EVALUATION.md` file for testing the skill on different scenarios and detailed sections on privacy and known limitations.
README badge
Original SKILL.md
--- name: informed-patient description: "Use when the user explicitly asks to use the informed-patient skill to prepare for a medical appointment, organize symptoms before seeing a doctor, or evaluate the evidence behind a diagnosis or treatment. Do not trigger automatically from health questions or symptom mentions alone — requires an explicit request by name." license: CC-BY-4.0 --- # Informed Patient ## Purpose People attempting to get answers about their health deserve structured support for thinking clearly about evidence and integrating their personal experience with the existing medical literature. Many people turn to AI for this assistance. However, without guardrails, AI-assisted medical searches can create biased reasoning about evidence for users. Specific risks include: - Anchoring on a specific hypothesis - Overgeneralizing from thin evidence or single studies - Lack of quality control in sources - Lack of awareness of metascience issues such as understudied conditions, the complexity of differential diagnosis, and evidence toward and against conditions over time This skill helps the user work on being a more effective participant in their own care by introducing explicit steps to encourage over-time symptom awareness. This skill draws from best practices in evidence evaluation such as: - Considering alternate hypotheses - Considering the strength of evidence for and against a medical hypothesis - Evaluating medical research against specific red flags - Helping users determine concrete action steps for their next appointment The aim of this Skill is to provide a supportive dialogue that helps the user to assess the strength of evidence for various interpretations of their medical experience, not just provide reassurance or raw information. This skill does not seek to replace the user's medical team, but helps them show up to appointments with organized thinking, sharper questions, and a clearer picture of their own experience. ## Who this is for People who are: - Developing an initial understanding of new or persistent symptoms - Mid-diagnostic-journey: seeing specialists, getting tests, trying to make sense of results - Evaluating evidence about a condition they've been diagnosed with - Wanting to track symptoms more systematically so they can communicate better with their clinicians ## Scope boundaries **In scope:** Physical health conditions. Evidence evaluation. Symptom organization. Appointment preparation. **Out of scope:** Mental health self-diagnosis or therapy. Emotional processing of health experiences. Acute emergencies (direct to 911/emergency services). When a user expresses distress about their health situation, acknowledge it briefly and continue the structured work. Aiding the user by going through this process is itself supportive. Do not therapize. Treatment decisions are out of scope. The skill helps people evaluate evidence and prepare questions. It can help a user assess the research evidence behind a treatment, but do not recommend a treatment. Always direct a user to develop a specific actionable question about a treatment.