> ## Documentation Index
> Fetch the complete documentation index at: https://docs.medera.info/llms.txt
> Use this file to discover all available pages before exploring further.
# Introduction to the Medera API
> Overview of the Medera application programming interfaces
Medera is the AI infrastructure stack for behavioral and mental health, built for clinical accuracy, evidence anchoring, and enterprise compliance. Behavioral health's specialized language, validated screeners, and crisis-detection requirements demand purpose-built AI infrastructure — not a generic LLM with a prompt. The Medera platform delivers multimodal sensing, medical-grade text generation, and agentic clinical automation through a single API that drops into existing EHR and payer workflows.
The platform runs on a HIPAA-grade runtime with Row-Level Security on every tenant-scoped table, AES-256 at rest for PHI, and 42 CFR Part 2 controls in the request path.
## Why choose Medera
| | |
| :-------------------------------------- | :--------------------------------------------------------------------------------------- |
| **Purpose-built for behavioral health** | Validated screeners, crisis detection, and 42 CFR Part 2 controls are first-class. |
| **Multimodal by default** | Voice, facial physiology, and linguistic content fuse into one defensible clinical read. |
| **Composable agents** | 24 production agents and core canvas nodes + clinical rails and a typed expert registry. |
| **Customizable and scalable** | Sandbox-first deployments with semver versioning and WORM audit on every promotion. |
Bespoke REST + WebSocket APIs JavaScript + Python SDKs Embeddable web components
Behavioral-health language proficiency Secure and compliant Real-time or asynchronous
***
## Integrate seamlessly
Medera integrates via REST + WebSocket.
The same API powers **Medera Studio** — a packaged, EHR-agnostic, real-time ambient scribe and intake agent. To embed Medera into an existing workflow or customize the interactions, dive into the API documentation below.
Voice + facial physiology + neurobehavioral fusion.
SOAP, referral, discharge, PA letters.
24 production agents + Agent Builder Canvas.
ICD-10, CPT, HCPCS, SNOMED, LOINC, RxNorm.
***
## Core capabilities
Real-time and asynchronous medical speech recognition, facial physiology, and vocal acoustics.
Reliable healthcare-focused LLM output, SOAP notes, and document generation.
24 composable clinical agents and a canvas to build your own.
***
## Two-service architecture
***
## Bringing it all together -
Each component has its own tab: [Multimodal Sensing](/multimodal/overview), [Text Generation](/textgen/overview), [Agentic Framework](/agentic/overview), and [Coding](/coding/overview).
* The [API Reference](/api-reference/welcome) tab provides detailed specifications for each endpoint.
* The [Console](https://medera.info/contact) is where you create accounts and Developer API Keys.
* Quickstarts outline how to get started fast. Start with [Authentication](/authentication/overview).
* The [SDKs](/sdks/overview) section covers JavaScript, Python, web components, and Postman.
* The [Examples Repository](https://github.com/medera-ai/examples) provides working code.
* The [Release Notes](/release-notes/overview) page provides the change policy and per-product notes.
If you have questions about implementing Medera in your environment or product, [contact us](https://medera.info/contact).