Quickstart
Go from zero to exploring a patient story in under 10 minutes.
Quickstart
By the end of this page you will have a workspace, a patient record ingested, and a resolved patient story you can browse.
What you need
- An email address or Google account (to create a free Clinia account)
- Node.js 20.9+ and pnpm 10+
Steps
Create a workspace
Go to console.clinia.cloud and sign up. After onboarding, create an organization, then create a workspace inside it. Your Workspace ID and Workspace URL are shown on the workspace details page.
Inside the workspace, scroll to OAuth Credentials, select New credential, and copy the Client ID and Client Secret immediately. The secret is shown only once.
See Provision your First Workspace for a detailed walkthrough.
Clone the Studio
The Context Engine Studio is a free, open-source app that connects to your workspace and gives you a visual interface to explore everything the engine produces.
git clone https://github.com/clinia/context-engine-studio.git
cd context-engine-studioConfigure and start
Create .env.local with your credentials:
cat > .env.local <<'EOF'
CLINIA_CONTEXT_ENGINE_API_URL=https://<workspace-id>.w.clinia.cloud
CLINIA_CONTEXT_ENGINE_OAUTH_CLIENT_ID=<your-client-id>
CLINIA_CONTEXT_ENGINE_OAUTH_CLIENT_SECRET=<your-client-secret>
EOFThen install and run:
pnpm install
pnpm devOpen http://localhost:3000.
Ingest a patient record
Download the Jeanne Tremblay synthetic dataset (FHIR R4 bundle + C-CDA documents for a synthetic 72-year-old patient with a decade of clinical history). Unzip it and drop the folder onto the Studio onboarding dropzone.
The engine ingests both sources, runs entity resolution across them, and builds the patient story. This takes a few seconds.
Browse the patient story
Once the ingest completes, expand the virtual file system tree in the sidebar:
/conditions/active/shows every condition the engine resolved/medications/active/shows the current medication regimen/encounters/lists encounters ordered in time
Select chronic_obstructive_lung_disease and open the Narrative tab. You are reading a condition story: a pre-assembled longitudinal narrative covering onset, treatments, labs, and complications, all built from the resolved graph.
Chat with the agent (optional)
This step requires an Anthropic API key with access to claude-opus-4-8. Add it to your .env.local:
echo 'ANTHROPIC_API_KEY=<your-anthropic-api-key>' >> .env.localRestart the Studio (pnpm dev), open the chat panel, and type:
Given the recent acute COPD exacerbation, is the current inhaler regimen appropriate and is bone health being monitored?
The agent browses the VFS, reads the relevant condition stories, and answers with citations to the files it used. Watch the tool calls to see exactly which paths it navigated. It discovers the ICS-bone-loss relationship from the condition stories without any application code that encodes that knowledge.