Documentation

The Context Engine for Health AI

Ingest fragmented patient records, resolve them into one patient graph, and give your agents the context they need over REST, MCP, and a virtual file system.

Get started API reference

Terminal

export API="https://<workspace-id>.w.clinia.cloud/v1"
export TOKEN="<access-token>"

# Send a FHIR R4 bundle for a patient
curl -X POST "$API/patients/demo-patient/ingestions/fhir/r4" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d @fhir-bundle.json

# Browse what the engine made of it
curl "$API/patients/demo-patient/vfs?path=/conditions/active" \
  -H "Authorization: Bearer $TOKEN"

Start with a quickstart

Two ways to your first patient story, about ten minutes each.

No code · 10 min

Explore a patient in Studio

Connect the open-source Studio to your workspace, ingest a synthetic patient, and browse the patient story it becomes.

Start the quickstart

Code · 10 min

Ingest and query with the API

Send a FHIR bundle and a CDA document, browse the virtual file system, and read a condition story.

Start the quickstart

Explore the documentation

Guides

Concepts, guides, and tutorials for the Context Engine

SDKs

Client libraries for the Context Engine API

Studio

The open-source app for exploring a workspace

API reference

REST operations, MCP tools, and VFS paths

Latest releasev0.6.0September 17, 2026

Trace anything back to where it came from. A typed graph exposes every entity the engine holds and every decision it made, and ingestion becomes a submission you can watch, account for, and read back.

Read the changelog