Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step. Self-hostable, runs with no API keys, MIT licensed. Ships an MCP server for Claude, Cursor and Continue.
Python agentic-aiai-securityalert-triageblue-teamcybersecuritydetection-engineeringincident-responselanggraphllm-agentsmcpmitre-attackmodel-context-protocolpurple-teamsecurity-operationsself-hostedsiemsoarsocthreat-detectionthreat-intelligence
id: llm-usage
name: AI / LLM Usage Audit Connector
version: "1.0.0"
plugin_type: connector
tier: stable
description: >
Pull organization audit logs from OpenAI or Anthropic — API key creation,
role grants, logging/MFA changes, project deletes. The governance surface for
AI adoption; pairs with the llm-* detection rules that fire on the emitted
event_type. Auth is an admin/audit-scoped API key.
author: AiSOC Core Team
tags: [saas, ai, llm, openai, anthropic, audit, connector]
homepage: https://github.com/beenuar/AiSOC/tree/main/services/connectors/app/connectors/llm_usage.py
license: MIT
min_aisoc_version: "7.0.0"
config_schema:
type: object
required: [provider, api_key]
properties:
provider: {type: string, enum: [openai, anthropic], description: Which provider's audit log to pull.}
api_key: {type: string, secret: true, description: Admin/audit-scoped API key.}