Plane 01 · Local
Operator runtime
Sessions, plans, tool traces, provenance, and workflow artifacts stay on the machine running OpenMed.
Terminal-native AI for clinical workflows that need control.
One inspectable agent for prior auth, appeals, coding, claims explanation, care coordination, consumer record imports, clinical documentation, FHIR work, literature search, and structured artifacts — across an inspectable operator runtime with protected hybrid medical services.
curl -fsSL "<install.sh URL shared during preview>" | bashLive execution
One prompt, one plan, visible execution.
§I
Runtime Boundary
Protected clinical services, inspectable runtime, explicit review before action.
OpenMed keeps the operator review loop visible, separates medical-service endpoints from the model-provider path, and routes sensitive or high-volume clinical processing to dedicated services when those paths are invoked.
Inspectable runtime
The TUI, sessions, plans, and generated artifacts remain visible and locally controlled.
Protected clinical services
Extraction, de-identification, terminology, LOINC, RxNorm, MedlinePlus, and HCC services run on dedicated medical-service endpoints when those paths are invoked.
Performance and cost efficiency
High-volume unstructured clinical processing can stay on dedicated service planes instead of consuming frontier-model context for every page.
Preview deployment
During preview, the native medical-service tier is provisioned as OpenMed-hosted infrastructure.
Workflow review
Deterministic workflows support draft preview, approval tokens, provenance, visible artifacts, and reviewer metadata on care-coordination finalization.
Opt-in audit trail
Private run traces can write to the operator's Hugging Face dataset, Hugging Face bucket, or local fallback with fingerprint-first payload handling.
Zero telemetry
No analytics, tracking, or background usage reporting built into the product.
§II
Deployment Model
Hybrid medical services, not a black-box healthcare prompt.
OpenMed separates the operator runtime, the model-provider path, and the medical service tier. Extraction, de-identification, terminology, LOINC, RxNorm, MedlinePlus, and HCC infrastructure stay visible, configurable, and better suited to sensitive or high-volume clinical processing than a single undifferentiated model path.
Plane 02 · Protected
Clinical extraction & de-identification
NER, PII detection, and de-identification run through a dedicated native service endpoint.
Plane 03 · Protected
Terminology, coding & HCC
PubMed, ICD-10, CPT, SNOMED, LOINC, RxNorm, MedlinePlus, HCC mapping, and RAF scoring run through a separate protected service plane.
§III
For Whom
Designed for the people shipping it and the teams using it.
OpenMed serves infrastructure-minded adopters and clinical operators in the same workflow surface.
For builders
Ship a terminal-native medical workflow with visible control surfaces.
An installable, configurable agent runtime instead of a black-box assistant — plans, traces, and artifacts you can read.
- Install with a provisioned shell command
- Configure models, profiles, native services, and optional MCP
- Inspect plans, tool traces, and workflow outputs
- Bring your own MCP endpoints when org tools are needed
For clinical teams
One interface for prior auth, coding, documentation, and care coordination.
Start from pre-built skills and built-in agent modes. Review workflow previews before finalizing.
- Start from pre-built clinical skills
- Choose the right built-in agent mode for the job
- Review workflow previews before finalizing
- Keep outputs structured, exportable, and evidence-linked
§IV
Capabilities
Capabilities grounded in the current product.
Everything on this page maps to workflows, tools, skills, or documentation already present in the repo. No vapor; if it's listed, it ships.
Clinical workflows
Prior Auth + Appeals
Review requests against CPT / ICD-10 criteria, LCD/NCD guidance, and denial evidence gaps with explicit recommendations.
cap.01
Care coordination
Inbox + Discharge
Prepare local inbox or discharge exports, run reviewer-safe triage and handoff workflows, and generate follow-up tasks plus patient-facing drafts.
cap.02
Consumer health
Record Imports + Summary
Import Apple Health, Health Connect, C-CDA, FHIR export, or labs files into one consumer-summary path with timeline, trends, reconciliation, questions, and narrative.
cap.03
Structured output
Coding + Claims
Run ICD-10, HCC, RAF, code-audit, and claims-explainer workflows with terminology validation and model-aware context when configured.
cap.04
Clinical documentation
Notes + SOAP
Turn visit transcripts into structured notes with vitals parsing, medication review flags, ICD-10 extraction, and E/M hints.
cap.05
FHIR + extraction
Structured FHIR Diff
Extract structured entities, build FHIR bundles, and compare them against a baseline bundle with visible diffs.
cap.06
Native medical tools
PubMed + Terminology
Use built-in PubMed, ICD-10, CPT, SNOMED, LOINC, RxNorm, MedlinePlus, HCC, RAF, and crosswalk tooling through configurable native service endpoints.
cap.07
Deployment
Hybrid Medical Services
Use hosted extraction, de-identification, terminology, education, and HCC services during preview, with a portable service boundary for later customer-managed deployment.
cap.08
Deterministic workflows
Draft, Finalize, Diff
Use draft preview, approval-token finalize, provenance, case-run history, and workflow diffing instead of opaque one-shot chat output.
cap.09
Audit trail
Private Trace Storage
Enable opt-in run traces to a private Hugging Face dataset, private storage bucket, or local directory with conservative fingerprint defaults.
cap.10
Remote integrations
Bring Your Own MCP
Attach remote JSON-RPC MCP servers when you need organization-specific tools beyond the native OpenMed surface.
cap.11
Trust
Mode Boundaries
Use
clinical, consumer, coordination, and plan modes to choose the right execution lane and review boundary for the job.cap.12
§V
Quickstart
Install, authenticate, ask.
Three steps from a fresh machine to a live workflow in the terminal across clinical, consumer, or care-coordination lanes.
01
Install
OpenMed is currently in preview. Installation is shared directly with approved evaluators.
Shared directly with approved evaluators
02
Authenticate
Use Codex OAuth or provide an API key, depending on the SDK path you want.
$
openmed login03
Ask
Describe the workflow in natural language and inspect the plan, tool traces, and outputs as they appear.
$
openmed§VI
Documentation
Docs that behave like product onboarding.
Start at the route you need instead of scanning one long index page.
Overview · Getting Started
Install, authenticate, choose a lane, and run the first workflow.
Product · CLI Reference
Understand the commands, agent modes, flags, and workflow-facing behavior.
Product · Agent TUI
See the terminal interface, plans, sessions, and keyboard flows.
Product · Skills
Review the built-in skill library and learn how to add your own local skills.
Product · Profiles
Inspect the built-in config presets and how they affect current settings.
Integrations · MCP Connectors
Add your own remote MCP servers and expose them as namespaced tools.
Integrations · Native Medical Services
See how extraction, de-identification, terminology, RxNorm, MedlinePlus, and HCC endpoints are deployed.
Trust · Privacy & Security
Understand runtime boundaries, PHI mode meaning, token storage, and network paths.
Trust · Audit Trail
Configure private run traces in Hugging Face datasets, Hugging Face buckets, or local fallback storage.
Start with the CLI and choose the lane you need.
Local sessions and artifacts, four built-in agent modes, configurable model and medical-service endpoints, optional MCP, and workflow outputs you can inspect before you depend on them. Consumer-import and care-coordination lanes remain local review flows with no outbound send or external-system writeback path.