AiSOC: Open-Source AI Security Operations Center Trends on GitHub
The open-source project AiSOC has gained significant traction on GitHub, adding over 5,700 stars this month for its self-hostable AI Security Operations Center. It features automated alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and native support for Model Context Protocol (MCP). This project bridges complex cybersecurity workflows with modern local AI capabilities, enabling teams to run automated threat investigations without relying on external API keys. Its integration with MCP allows security professionals to leverage tools like Claude and Cursor directly within their workflows. Written in Python and distributed under an MIT license, AiSOC includes a replayable decision ledger that logs every step taken by the AI agent for auditing. It also ships an MCP server designed to integrate seamlessly with environments like Claude, Cursor, and Continue.
## BACKGROUND
A Security Operations Center (SOC) is a centralized unit that deals with security issues on an organization's systems, handling high volumes of alerts daily. Traditional triage requires manual filtering of these alerts, whereas autonomous LLM agents can reason through evidence and investigate threats using frameworks like MITRE ATT&CK.