Integration

Infermedica integration

Infermedica is a medical symptom-assessment and triage engine. Find symptoms, risk factors and conditions by phrase or in free text, then run a symptom-checker interview: run_diagnosis returns the next question and a ranking of probable conditions, and run_triage, recommend_specialist, explain_condition and get_question_rationale turn the same evidence into an urgency level, a specialist and channel, and reasons. It also lists and describes every medical concept, returns patient education articles (a paid add-on) and records patient intent surveys. The cardinal-features route, absent from the vendor guides, is not exposed. Authentication uses two headers, App-Id and App-Key, from the Apps page of the Infermedica developer portal; trial plans limit calls to suggest, triage, explain and rationale.

Connect Infermedica How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Infermedica.Built from the 19 Infermedica tools below, on the connection you add once.

What you can do with Infermedica

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Concept, 2 operations

get_conceptGet conceptGETReturns one medical concept (a condition, symptom or risk factor) by its id, with its type, name and plain-language common name.
list_conceptsList conceptsGETLists medical concepts (conditions, symptoms and risk factors) with their id, name and plain-language common name. Filter by type or by a comma-separated list of ids.

Condition, 2 operations

get_conditionGet conditionGETReturns one condition for a given patient age: name, common name, categories, prevalence, acuteness, severity, triage level, recommended channel, ICD-10 code and a short advice hint.
list_conditionsList conditionsGETLists every condition available for a given patient age, with its categories, prevalence, acuteness, severity, triage level, ICD-10 code and advice hint.

Risk factor, 2 operations

get_risk_factorGet risk factorGETReturns one risk factor for a given patient age, including the question to ask the patient about it in first and third person.
list_risk_factorsList risk factorsGETLists every risk factor (for example smoking, high BMI, recent travel) available for a given patient age, with its plain-language name, seriousness and category.

Symptom, 2 operations

get_symptomGet symptomGETReturns one symptom for a given patient age, with the question to ask the patient about it, its parent symptom and its more specific child symptoms.
list_symptomsList symptomsGETLists every symptom available for a given patient age, with its plain-language name, seriousness, parent symptom and child symptoms.

Diagnosi, 1 operation

run_diagnosisRun diagnosisPOSTRuns one step of a symptom-checker interview: from sex, age and the evidence so far, returns the next question to ask, a ranking of probable conditions with probabilities, and should_stop once enough is known.

Explain, 1 operation

explain_conditionExplain conditionPOSTExplains one condition of a diagnosis ranking: which reported observations support it, which conflict with it (reported absent) and which are unconfirmed. Takes the same patient data as run_diagnosis plus the condition id; at most 60 evidence items.

Feedback, 1 operation

submit_intent_surveySubmit intent surveyPOSTRecords an intent survey for an interview: the kind of care the patient planned before the interview and after seeing the triage recommendation. Send the same Interview-Id as the interview it belongs to.

Info, 1 operation

get_model_infoGet model infoGETReturns the API version and the medical content model in use: when it was last updated and how many conditions, symptoms and risk factors it holds. Check updated_at before reusing stored concept ids.

Parse, 1 operation

parse_text_for_symptomsParse text for symptomsPOSTFinds symptoms and risk factors mentioned in free text (English, at most 2,048 characters), including negated ones and misspellings. Each mention comes with an id and a choice_id (present or absent) ready to use as evidence.

Patient education, 1 operation

get_patient_educationGet patient educationGETReturns the patient education article for a condition, as rich-text sections (summary, causes, diagnosis, home treatment, prevention, when to see a professional, sources). Needs the interview_token from run_diagnosis; a paid add-on.

Rationale, 1 operation

get_question_rationaleGet question rationalePOSTExplains why the interview asks its current question: a rationale code (r0 to r6) and the observations and conditions behind it. Takes the same patient data as run_diagnosis.

Recommend specialist, 1 operation

recommend_specialistRecommend specialistPOSTRecommends the kind of specialist (for example a general practitioner or a neurologist) and the consultation channel for a case. Takes the same patient data as run_diagnosis; not available for emergency triage levels.

Search, 1 operation

search_conceptsSearch conceptsGETFinds symptoms, risk factors or conditions matching a phrase, even a few typed letters, for autocomplete. Returns each matching concept id with the label that matched (its name or a synonym).

Suggest, 1 operation

suggest_related_observationsSuggest related observationsPOSTSuggests observations related to the patient's evidence (symptoms often reported together, demographic or evidence-based risk factors, or red flags), to ask before starting run_diagnosis.

Triage, 1 operation

run_triageRun triagePOSTClassifies the urgency of a case from the same patient data as run_diagnosis: a triage level from self_care to emergency_ambulance, the serious observations reported, and the root cause of that level.

Use Infermedica from Claude Code, Cursor or Codex

Connect Infermedica in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Infermedica operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Infermedica

Create your workspace
Sign up free. Nothing to install.
Connect Infermedica once
Add your Infermedica connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.