OpenMed Agent

Preview medical workflow agent for the terminal — local operator runtime, protected medical services, reviewable outputs

OpenMed Agent is currently in preview. It gives clinicians, healthcare operators, and technical teams a terminal-native workspace for clinical and operational workflows such as prior authorization review, appeal review, coding audit, consumer health summaries, care coordination, and clinical documentation.

---

Install

curl -fsSL "<install.sh URL shared during preview>" | bash

---

Preview Status

OpenMed Agent is in preview, which means:

• the product surface is real and usable today
• the hosted medical-service tier is still evolving
• some service-backed capabilities are operated by OpenMed during preview rather than fully self-serve
• workflow, deployment, and integration details may continue to tighten as the product hardens

During preview, OpenMed operates the protected clinical-service endpoints so evaluators do not need to deploy extraction and terminology infrastructure themselves.

---

What It Does

OpenMed Agent combines LLM reasoning with deterministic workflows and native medical tools. You describe what you need in natural language, and the agent can:

• review prior authorization and appeal cases against structured criteria
• audit ICD-10 coding with HCC and RAF context
• explain EOB and claims data in plain language
• extract entities or de-identify clinical text
• summarize consumer health records from Apple Health, Health Connect export, C-CDA, FHIR export, and labs files
• triage inbox threads, draft reviewer-safe replies, and generate discharge handoffs
• search PubMed and use protected terminology services for ICD-10, CPT, SNOMED, LOINC, RxNorm, MedlinePlus, HCC, and RAF

openmed            # Launch the agent
openmed agent      # Launch with options (model, reasoning effort, skill, agent mode)
openmed --help     # Full CLI reference

---

What Ships In Preview

• 62 built-in native tools
• 13 deterministic workflows with draft/finalize
• 13 built-in skills
• 4 agent modes: clinical, consumer, coordination, plan
• 104 demo scenarios covering all capability areas

---

Key Capabilities

Protected Medical Services

• Clinical extraction — entity extraction, PII detection, and de-identification through protected native service endpoints
• Medical terminology and coding — ICD-10, CPT, SNOMED, LOINC, RxNorm, MedlinePlus, HCC, RAF, validation, crosswalks, and PubMed-backed lookup
• Configurable deployment boundary — the operator runtime stays local while medical-service endpoints can be moved across hosted, cloud, or customer-managed environments

Clinical And Operational Workflows

• Prior authorization and appeals — review requests against deterministic criteria and structured evidence
• Coding audit — specificity review, compliance flags, HCC mapping, and RAF impact
• Claims explanation — patient-friendly EOB and billing explanations with next-step guidance
• Clinical documentation — structured SOAP-style documentation from notes or transcripts
• Care coordination — inbox triage, reviewer-safe patient drafts, discharge handoff, PCP handoff, and follow-up tasks
• Consumer health — imported health-record normalization, timeline/trend analysis, visit-prep questions, narratives, reconciliation, optional education topics, and optional FHIR output

Agent Runtime

• Project instructions — drop an OPENMED.md file in your project root to customize agent behavior per workspace
• Permission policy — rule-based auto/acceptall/denyall/plan modes with per-tool allow/deny/ask rules from project or user settings
• Tool safety classification — every tool carries is_read_only, is_concurrent_safe, and is_destructive metadata used by the permission system
• Oversized result handling — large tool outputs automatically persisted to disk with compact in-context stubs
• Runtime diagnostics — /config command shows effective settings, loaded sources, and active project instructions

Agent Experience

• Interactive TUI — terminal interface with sessions, themes, model switching, skill switching, and workflow execution
• Draft/finalize workflow lifecycle — reviewable cards, artifacts, provenance, and workflow diffs
• Skill system — built-in clinical skills that bias how the agent approaches domain-specific work
• Plan auto-advance — structured plans with deterministic tool-based progress tracking
• Session persistence — save, restore, fork, and rollback conversation sessions
• Optional MCP integration — connect external medical or institutional systems without changing the native tool surface
• Self-update — built-in binary update path with release checks
• No telemetry — OpenMed does not ship built-in analytics or phone-home tracking

---

Important Boundary Notes

• OpenMed Agent is not a universally offline product.
• The operator runtime is local, but many medical capabilities call configured protected endpoints when those paths are invoked.
• PHI handling modes are operator-visible workflow settings, not a blanket guarantee that every code path is automatically enforced the same way.
• Reviewability is a core product characteristic. Final clinical artifacts and care-coordination outputs are designed to be inspected before use.

---

Visit agent.openmed.life for full documentation.

---

Website | Documentation | X/Twitter | LinkedIn