AI-powered mental health support platform with anonymous chat, crisis detection, and peer support - Self-hosted & open source by Team ZenYukti
AI safety benchmark for long-term caregiving relationships. Tests crisis detection, regulatory compliance, and care quality across multi-turn conversations. Includes GiveCare system paper and InvisibleBench evaluation framework.
Accessible therapeutic AI optimized for Apple Silicon (M1/M2/M4) with 90% crisis detection accuracy. Implementation of ChatThero paper with professional-grade evaluation, Docker support, and comprehensive safety features. Runs locally with 4-6GB RAM.
Production-Grade Multi-Agent Therapeutic AI Platform | 17 Microservices | 100% Crisis Detection | HIPAA Compliant
Two complementary AI agent skills for comprehensive mental health support: Crisis Detection for safety monitoring, and Praise Group for positive reinforcement.
VoiceMind: A voice-first AI mental health companion. Flutter + FastAPI + Google ADK with dual-channel emotion fusion, tiered crisis detection, and offline coping engine.
Mental health support platform with PHQ-9/GAD-7 assessments, AI analysis, and 24/7 WhatsApp chatbot. HIPAA-compliant design with crisis detection.
OVAL — RAG-powered brand intelligence platform. Monitors 8 platforms, AI-classifies mentions for sentiment/severity, auto-routes actions, and detects crises 6 hours before they hit Google. Built with Python, Next.js 14, Supabase, and Claude/GPT-4o.
MindMate: AI Mental Health Chatbot by Mental Health Innovators for Hack-o-Spider. Offers 24/7 support, privacy, and AI empathy (OpenAI, Gemini, Claude) with crisis detection and fast responses via FastAPI, Streamlit, Docker.
Serenity-Mental-Health-Chatbot: Serenity is a mental health chatbot built with a BiLSTM intent model, RAG system, and crisis detection. It offers empathetic conversations, safety handling, and a modern UI for real-time emotional support, analytics, and personalized user interaction. github.com
Automated time series-based anomaly detection framework for crisis monitoring. Detects unusual behaviors, spikes, and trends for early warning systems in finance, industry, and cyber-security.
Binary safety verdicts (SAFE/HELD/LEAK/MISS/BROKE) + persona fan-out for LLM pipeline evals
Add a description, image, and links to the crisis-detection topic page so that developers can more easily learn about it.
To associate your repository with the crisis-detection topic, visit your repo's landing page and select "manage topics."