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: det-application-016
name: Push to Protected Branch by Non-Admin
description: AiSOC v1 curated detection. Triggers on the application signal described by 'Push to Protected
Branch by Non-Admin'. Watch the 1 documented false-positive case before tuning.
version: 1.0.0
severity: high
tags:
- mitre.attack.t1195.002
- tlp.white
category: application
log_source:
product: github
service: audit
detection:
condition: |
action == "git.push"
AND ref_in_protected_branches == true
AND actor_role NOT IN ["admin", "release"]
false_positives:
- Emergency hotfix workflow (must reference change ticket)
playbook: tpl-supply-chain
enabled: false
quarantine_reason: 'not executable (needs-connector-field): wants `ref_in_protected_branches` and `actor_role`
on a git.push, and the audit entry carries neither a ref nor the actor''s org role. The observable form
of a protection bypass in this telemetry is `protected_branch.policy_override`, which app-protected-branch-disabled
already matches.'
author: AiSOC
created: '2026-05-03'
modified: '2026-05-03'