lynox
Run your business, not your tools — AI runtime for CRM, workflows, research, monitoring. Self-hosted or managed hosting.
Ficha actualizada 22 sept 2026, 19:22 · upstream 22 sept 2026, 19:13 · release v2.14.2
Qué es lynox
Descripción editorial construida a partir del repositorio, su documentación y sus señales de mantenimiento.
One source-available (ELv2) agent that learns your business context and executes tasks across multiple domains. Lynox connects any API, runs your workflows, researches the web, monitors for changes, and maintains a persistent knowledge graph of your entities and relationships.
It is designed to be self-hosted with no vendor lock-in, offering full control over data, models, and infrastructure. Whether you need to summarize PDFs, research topics, automate daily tasks, or monitor competitor activity, lynox provides a unified interface for orchestrating intelligent workflows.
The platform supports multiple LLM providers including Anthropic, Mistral, Ollama, Fireworks AI, vLLM, LocalAI, Groq, and others via preset configurations. It features an interactive Web UI with real-time tool call streaming, activity bars showing sub-phase progress, and a mobile PWA for on-the-go access.
Key capabilities include knowledge graph construction from your data, entity resolution across documents, background worker scheduling, Google Workspace integration for calendar and file management, process capture via workflow templates, and secure AES-256 encrypted vaults. Lynox is built by Rafael Burlet in Rapperswil SG with a focus on practical business value over theoretical abstractions.
- Summarize PDF documents into structured key points for quick review
- Research specific topics and synthesize findings across multiple sources
- Automate daily tasks like checking emails, scheduling meetings, or updating calendars
- Monitor competitor websites and track changes in real-time
- Create reusable workflow templates that can be scheduled and reused across projects
- Build a personal knowledge base from scattered notes, documents, and web searches
- Self-hosted with no vendor lock-in — you own your data and models entirely
- Real-time tool call streaming via activity bars eliminates long waits between requests
- Supports multiple LLM providers including Anthropic, Mistral, Ollama, Fireworks AI, vLLM, LocalAI, Groq, and others
- Voice input/output is functional but not polished — iOS Safari has Web Audio quirks requiring Chrome
- Native calendar integration (CalDAV/ICS) is in development; currently uses Google Workspace OAuth instead
- No multi-user or team accounts supported — single instance per user with one vault and knowledge graph
Perfil de infraestructura
Estimación inicial derivada del repositorio y de los servicios detectados.
Despliegue detectado
Qué señales reales de instalación hemos encontrado.
Install via Docker Compose: copy .env.example to .env, add your API key, run `docker compose up`. For production, enable AES-256-GCM encryption in the environment variables. The Web UI runs on port 3000 and can be accessed from any device with HTTPS enabled. Use the interactive installer for a guided setup experience that prompts provider selection, API key generation, Docker Compose scaffolding, container pull, health checks, and final startup. For self-hosted deployments, ensure Node.js 22+ is installed and consider using `better-sqlite3` which requires C++ build tools on macOS (xcode-select --install) or Ubuntu (sudo apt-get install build-essential python3).
Stack y dependencias
Componentes inferidos a partir de manifests, configuración y documentación.
Puede sustituir a
Relaciones detectadas por el análisis del repositorio y su categoría.
Experiencia real de la comunidad
Ayuda a contrastar la estimación automática con instalaciones reales.
Todavía no hay datos comunitarios para este proyecto.