This repository hosts SmartSpectra SDK from PresageTech for measuring vitals such as pulse, breathing, and more using a camera. The SDK supports multiple platforms, including Android, iOS, and C++ for Windows, Mac, and Linux.
To get started, follow the instructions for one of our currently supported platforms. Each platform has its own integration guide and example applications to help you get up and running quickly. Documentation is available at https://smartspectra.presagetech.com/.
This repo ships using-smartspectra, an Agent Skill that teaches an AI coding assistant how to build apps with the SDK. The same skill works in both Claude Code and OpenAI Codex:
- Claude Code:
/plugin marketplace add Presage-Security/SmartSpectra, then/plugin install smartspectra-sdk@smartspectra - Codex:
$skill-installer Presage-Security/SmartSpectra/.agents/skills/using-smartspectra
Then ask your assistant to build with SmartSpectra. Full guide: docs/agent-skill.md.
Presage also hosts an MCP (Model Context Protocol) server at https://mcp.presagetech.com/mcp. Connect it and your assistant can work with your developer account directly — fetch or rotate your API key, register an iOS or Android app ID, download its OAuth config file, and check your plan and credits. Setup: docs/mcp-server.md.
-
Cardiac Waveform
Real-time pulse pleth waveform supporting calculation of pulse rate and heart rate variability. -
Breathing Waveform
Real-time breathing waveform supporting biofeedback and breathing rate. -
Myofacial Analysis
Supporting face-point analysis, iris tracking, blinking detection, talking detection, and facial expression classification. -
Relative Blood Pressure Waveform
Relative blood pressure waveform shape. -
Integrated Quality Control
Confidence and stability metrics providing insight into the confidence in the signal. User feedback on imaging conditions to support successful use. -
Camera Selection
Front or rear facing camera selection on iOS or Android and specification of camera input for applications using the C++ SDK. -
LLM Insights
Turn the vitals the SDK computes on-device into natural-language analysis from a large language model, on request. See the LLM Insights overview.
Model cards include:
- Relative arterial pressure waveform
- Heart rate variability
- Pulse rate
- Face analysis
- Breathing rate
- Upper and lower breathing waveforms
Review the SmartSpectra model cards and limitations for metric-specific operating ranges, validation notes, and known limitations.
The SDK can report a small, aggregate, per-session diagnostic summary so we can measure release quality across the devices and conditions the SDK runs on. It is aggregate-only — no raw video, no measured vitals values, no user or device identifiers — and it is opt-out. See SDK Telemetry & Privacy for exactly what is and isn't collected and how to disable it.
- We support API key authentication for C++, iOS, and Android. We also support OAuth authentication for iOS and Android. See the platform-specific guides for setup instructions.
- Either path can be set up by an AI assistant instead of by hand — see docs/mcp-server.md.
For Android integration, refer to the Android README. The guide includes:
- Prerequisites and setup instructions.
- Maven setup for stable releases, release candidates, and snapshots.
- Integration steps for your app.
- Example usage and troubleshooting tips.
For iOS integration, refer to the iOS README. The guide includes:
- Prerequisites and setup instructions.
- Integration steps for your app using Swift Package Manager.
- Example usage and troubleshooting tips.
For C++ integration on Windows, macOS, and Linux, refer to the C++ README. The guide includes:
- Supported systems and architectures.
- Installation via the prebuilt ZIP (Windows), Homebrew (macOS), or apt (Linux).
- Build instructions and example applications.
- A Linux redistribution path for bundling the published SDK tarball into your
own
.deb: Redistribute SmartSpectra on Linux.
For additional support, contact support@presagetech.com or submit a GitHub issue.