Cortex

•
4
•

Aggregates homelab syslog, Docker, OTLP, and AI-session logs into SQLite for full-text search, providing intelligence and analysis via MCP, REST, and CLI interfaces.

Cortex is a Rust-based log aggregation and intelligence platform specifically designed for homelab environments. It centralizes various log sources, including syslog (UDP/TCP), Docker logs, OTLP ingest, managed file-tail, inventory, heartbeat, and AI-session signals, storing them efficiently in SQLite with FTS5 full-text indexing. The platform offers a rich set of capabilities through its Model Context Protocol (MCP), REST API, and CLI adapters, enabling action-based log search, filtering, timeline generation, error analysis, inventory management, and graph-backed topology investigations. A key feature is its ability to correlate AI transcript sessions with non-AI logs, facilitating advanced intelligence and incident evidence bundle generation within a self-hosted context. Cortex focuses on bounded investigation actions and explicit admin gates, making it a robust tool for homelab observability without aiming to replace a full SIEM.

Características Principales

01UDP and TCP syslog receiver on port 1514
02Ingestion of Docker logs, OTLP, managed file-tail, inventory, heartbeat, and AI-session signals
032 GitHub stars
04SQLite + FTS5 log storage with retention and storage-budget controls
05MCP, REST, and CLI adapters for comprehensive log search, analysis, and operations
06AI transcript correlation and derived graph projection tables for topology and investigation

Casos de Uso

01Centralized log aggregation and full-text search for self-hosted homelab infrastructure
02Correlating AI project transcripts with system and application logs for contextual insights
03Incident investigation, error monitoring, and generating evidence bundles from diverse log data