Skip to content

Abby

Abby is a patient-state agent built around the Abridge workflow.

Core trick

After an ambient visit, an agent verifies the patient by OTP, runs a specialty check-in, and queues actions for clinician approval before write-back.

Tags

Description

Abby is a patient-state agent built around the Abridge workflow. After an ambient clinical note is created, Abby continues the conversation outside the visit: it reads transcript/FHIR context, securely verifies the patient, performs a focused specialty-specific check-in, extracts interval changes and red flags, and prepares an action-oriented brief for the clinician. For this demo, Abby uses the synthetic Abridge ambient/FHIR dataset to show a full loop: admin and provider setup, patient outreach, OTP-style identity verification, patient chat, structured FHIR-shaped patient-reported updates, safety/evaluation gates, and clinician approval before any write-back or scheduling action. The goal is to turn Abridge from a note-generation endpoint into the beginning of an ongoing care loop: what changed, what matters, what needs clinician attention, and what can be safely queued for follow-up.
aalami5/abby★ 00 forksTypeScriptCreated Jul 18, 2026 · last push Aug 13, 2026

README excerpt

# Abby

Abby is a local hackathon prototype for a patient-state agent that sits around Abridge workflows.

The app uses the synthetic Abridge ambient/FHIR dataset to demonstrate:

- admin/operator outreach setup
- admin, provider, and patient directory setup
- phone-number OTP authentication flow with Twilio Verify-ready endpoints
- transcript-grounded patient intake
- provider Action Briefs
- FHIR-shaped patient-generated data
- mocked tool-connected workflows
- an eval harness for safety and quality gates
- clinician approval gating before write-back or scheduling
- server-side run lifecycle endpoints with browser fallback
- FHIR Bundle JSON export

## App Instructions

The provider-facing Abby instructions for the demo live in:

```text
ABBY_INSTRUCTIONS.md
```

That file is intentionally kept at the repository root so it is easy to identify in GitHub.

The same source is clearly identified in the directory database for agent lookup:

- Document: `directory-v1`
- Reference catalog: `agentInstructionReferences`
- Reference ID: `abby-app-instructions`
- Provider record: `person-oliver-aalami`
- Instruction field: `abbyInstructions`
- Source metadata: `abbyInstructionsSourceFile`, `abbyInstructionsSourcePath`, `abbyInstructionsSourceUrl`

## Run

```bash
npm install
npm run dev
```

Open `http://127.0.0.1:5173/`.

## Cloud

Current Vercel deployment:

```text
https://abby-mocha.vercel.app
```

Health and capability endpoints:

```text
https://abby-mocha.vercel.app/api/health

Team