内容
# OpenMed Agent
> Clinical AI in your terminal — medical workflows, coding and terminology tools, and reviewable output.
OpenMed Agent helps clinicians, healthcare operators, and technical teams work with clinical records, prior authorizations, appeals, coding, documentation, and care coordination. Choose a model provider, describe the task, and review the resulting plans, citations, and workflow artifacts.
**Latest release: [v0.5.0](https://github.com/openmed-labs/openmed-agent/releases/tag/v0.5.0)** · [All release notes](https://github.com/openmed-labs/openmed-agent/releases) · [Documentation](https://agent.openmed.life/docs/)
OpenMed Agent is in preview. This repository provides product information and public release notes. Installation instructions are shared with approved evaluators through the [preview access page](https://agent.openmed.life/).
## What's new
- **Doubleword real-time:** stream responses and use agent tools in the TUI with `openmed agent --provider doubleword`.
- **Async flex:** submit individual background requests, close the terminal, and retrieve the results later using the saved ID. On-demand priority requests are also available through the CLI.
- **24-hour batches:** validate and submit independent JSONL requests, retrieve available results, resume waiting, or request cancellation. Use this for backlogs and overnight work that can trade latency for lower inference cost.
- **Clearer model selection:** eleven Doubleword generation models are included, with a configurable list, provider labels in search results, and context limits for automatic compaction.
All three modes share one key. The command selects the mode: the TUI is real-time, while async and batch jobs use the CLI. Direct request files do not execute agent tools or automatically de-identify input.
See the [v0.5.0 notes](https://github.com/openmed-labs/openmed-agent/releases/tag/v0.5.0), the [three-mode quickstart](https://agent.openmed.life/docs/providers/), and the [release history](https://github.com/openmed-labs/openmed-agent/releases).
## Get started
Use the installation instructions supplied with your preview access, then run:
```bash
openmed login # Sign in with ChatGPT
openmed # Open the terminal workspace
openmed --models # List available models
openmed --help # Show CLI commands
```
You can also configure your own OpenAI, Anthropic, OpenRouter, MiMo, or Doubleword API key, or connect a custom OpenAI-compatible server. Follow the [provider setup guide](https://agent.openmed.life/docs/providers/).
GPT-5.6 Terra at medium reasoning is the default for this preview. Switch the model, reasoning effort, skill, and agent mode from the terminal interface.
To check for or install an update:
```bash
openmed --check-update
openmed update
```
[Getting started](https://agent.openmed.life/docs/getting-started/) · [CLI reference](https://agent.openmed.life/docs/cli/) · [Terminal interface](https://agent.openmed.life/docs/tui/)
## Clinical and operational capabilities
| Area | What you can do |
| --- | --- |
| Prior authorization and appeals | Review requests, coverage criteria, and denial evidence; prepare reviewable drafts. |
| Coding and claims | Audit ICD-10 coding, inspect HCC and RAF context, use clinical classification and risk-model tools, and explain EOB or claims data. |
| Clinical documentation | Turn notes or transcripts into structured clinical documentation and SOAP-style drafts. |
| Care coordination | Triage inbox threads, prepare patient replies, create discharge and PCP handoffs, and organize follow-up tasks. |
| Consumer health | Summarize Apple Health, Health Connect, C-CDA, FHIR exports, and lab files; inspect timelines, trends, and visit-preparation questions. |
| Clinical text and evidence | Extract entities, de-identify text, search PubMed, and use coding and terminology lookups. |
Medical tools cover ICD-10, CPT, SNOMED CT, LOINC, RxNorm, MedlinePlus, HCC, CCSR, comorbidities, RxHCC, ESRD, and Orphanet. Availability depends on the medical services configured for your installation.
## Work in a reviewable workspace
- **Draft and finalize:** inspect workflow previews, artifacts, citations, and differences before completing a workflow.
- **Saved sessions:** return to conversations with their plans and previous tool results available for follow-up work.
- **Files and citations:** use `@` file and directory completion, select response text, and copy code blocks with `/copy`.
- **Project instructions and skills:** use `OPENMED.md` for workspace guidance and select built-in or custom skills for domain work.
- **Permissions:** configure tool approval behavior for your workspace.
- **Optional MCP:** connect additional medical or institutional tools.
- **Optional web search:** enable client-side search when needed; it is off by default.
[Skills](https://agent.openmed.life/docs/skills/) · [Configuration](https://agent.openmed.life/docs/configuration/) · [Web search](https://agent.openmed.life/docs/web-search/)
## Preview and data handling
Sessions and generated artifacts are stored on the machine running OpenMed. Model requests and medical-service calls use your configured providers and endpoints. Choosing a local model does not make connected medical services offline.
During preview, some medical services are operated by OpenMed. Service availability and integration details may change as the preview develops. PHI handling modes are workflow settings; they do not provide a blanket privacy guarantee. Review clinical outputs before use.
[Native medical services](https://agent.openmed.life/docs/native-medical-services/) · [Privacy and security](https://agent.openmed.life/docs/privacy/)
---
[Website](https://agent.openmed.life/) · [Documentation](https://agent.openmed.life/docs/) · [Releases](https://github.com/openmed-labs/openmed-agent/releases) · [Repository license](LICENSE)
MCP 市场