Products/AI Agent Compliance / Audit Logging / Healthcare AI/Cyris

Cyris

The system of record for AI agent decisions

AI Agent Compliance / Audit Logging / Healthcare AIai-complianceaudit-logginghealthcareanomaly-detectionReviewed

Our Take

Cyris is building the paper trail for AI decisions that regulators are eventually going to demand — every model output tagged with the inputs, logic, and context that made it happen. Given how fast AI governance laws are moving in the EU and stateside, this feels like a "build the fence before the horse runs off" situation. The value prop is straightforward enough that I don't need to overexplain it, but I'd need to see actual trace examples to tell you if they've nailed the integration piece or just the concept.

Provides audit logging, anomaly detection, and compliance tracking for AI agents used in healthcare settings. Wraps agents in two lines of code to capture every decision as audit-ready evidence.

Key Features

Per-decision audit log with hash-chained input, output, agent, timestamp, Real-time anomaly feed with severity classification, Patient-level workflow trace, end to end, Incident history with downstream impact tracked, Drift metrics per agent, exportable per window, PHI detection before it leaves your process, Auto-instrumented - no code changes required, Agent discovery from live traffic, Workflow graph mapping from real traffic, Live compliance trust URL for hospital teams, HIPAA, SOX, EU AI Act compliance support, Cost tracking per agent and model, BAA (Business Associate Agreement) support

Problem It Solves

Hospitals require AI vendors to prove explainability, anomaly detection, and data lineage before signing contracts - most startups cannot meet these security review requirements.

Target Customer

AI-native healthcare startups in active sales conversations with hospitals

Use Cases

Hospital security reviews, Healthcare AI compliance audits, Patient outcome tracing, Incident response, Model drift monitoring, AI decision provenance

Pricing Details

Pricing not explicitly displayed on page

Differentiator

Two lines of code to instrument all agents - auto-discovers agents from live traffic and draws workflow graphs automatically within 7 days. Provides live trust URL for compliance teams instead of static PDFs.

Why Now

Hospitals now require AI vendors to prove explainability, anomaly detection, and data lineage before signing - this is a new compliance barrier for AI startups.

Traction

Customers Mentioned: Northstar Health AI · Notable Metrics: 94% audit readiness; 1,284,902 total decisions on the record; 127 days continuously logging; 8,412 PHI events logged; 2 anomalies in last 30 days; BAA on file

Key Facts

The people behind Cyris

Cyris Kissane

profile

CEO & Co-Founder

CEO & Co-Founder of Cyris (AI agent audit trail for healthcare). Backed by Convoi Ventures. Previously Google Fuchsia Tool Chain Team (2022). AI researcher with proprietary Mamba/DeepSeek precursor work.

Links

Website Source: product-hunt

Want products like this in your inbox every morning?