phi-redaction

Open-source guardrails between AI agents and FHIR clinical data — PHI redaction, immutable audit, step-up auth, tenant isolation. MCP server + OpenAI/Gemini adapters. A healthclaw.io project.

PDF redaction that runs entirely on your device. Five detection layers including Gemma 4 contextual PHI. No uploads, no servers, no leaks.

Open-source MCP servers for clinical decision support — calculators, drugs, evidence, terminologies. Local-first, PHI-aware, fail-loud.

DRAFT: This project synchronizes SysAid support tickets (Service Requests) into Azure DevOps (ADO) as Bugs

Veterinary discharge note generator with multi-provider LLM support (OpenAI, Anthropic), PHI redaction guardrails, FastAPI REST API, and Docker deployment. Built with Python 3.13, Jinja2 templating, and comprehensive test suite.

NPS and sentiment intelligence for medical AI companions. Features PHI-safe redaction, automated clinical safety gating, on-call paging, and a clinical-trust dashboard.

Security-hardened clinical AI gateway with PHI redaction, prompt injection defense, audit logging, and local LLM inference.

Governed long-term memory for clinical AI agents — a FastAPI service on mem0 with PHI redaction, patient scoping, and an audit trail

Governed LLM gateway — PHI guardrails (redact/block), multi-provider routing with fallback, and per-call cost accounting. OpenAI-compatible, built on LiteLLM