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: snowflake
name: Snowflake Account Activity Connector
version: "1.0.0"
plugin_type: connector
tier: stable
description: >
Snowflake audit + account-usage connector. Distinct from the existing
snowflake-events plugin (which is the reference Python + Go SDK
implementation under plugins/snowflake-events/): this is the
first-party AiSOC core connector that ships with the platform. Polls
the SNOWFLAKE.ACCOUNT_USAGE schema for LOGIN_HISTORY and QUERY_HISTORY
views, normalises rows to AiSOC alert shape, and emits anomaly
candidates (failed login bursts, large-result-set downloads, queries
from unexpected regions, service-account queries from human-named
warehouses).
author: AiSOC Core Team
tags: [data, snowflake, warehouse, exfil, audit, connector]
homepage: https://github.com/beenuar/AiSOC/tree/main/services/connectors/app/connectors/snowflake.py
license: MIT
min_aisoc_version: "4.0.0"
config_schema:
type: object
required: [account, user, password, warehouse]
properties:
account:
type: string
description: >
Snowflake account identifier (e.g. xy12345.us-east-1). This is
the same value that prefixes your Snowflake URL.
user:
type: string
description: >
Snowflake user with read access to SNOWFLAKE.ACCOUNT_USAGE. We
recommend creating a dedicated AISOC_AUDIT_READER user with
only that schema's USAGE + SELECT grants.
password:
type: string
description: >
Password or programmatic access token for the user above.
Stored in the AiSOC credential vault and decrypted only at
connect time.
secret: true
warehouse:
type: string
description: >
Warehouse used to execute the audit queries. Use a dedicated
X-Small warehouse so audit-poll cost is predictable and does
not contend with production analytical workloads.
role:
type: string
default: ACCOUNTADMIN
description: >
Role with access to SNOWFLAKE.ACCOUNT_USAGE. ACCOUNTADMIN is
the default in Snowflake but you can substitute a custom role
with the SHOW_PRIVILEGES or MONITOR_USAGE system privileges.
poll_interval_seconds:
type: integer
default: 300
minimum: 60
maximum: 3600
description: >
How often the connector polls. ACCOUNT_USAGE views have a
latency of up to 45 minutes from real-time per Snowflake docs,
so going below 60s gives no detection-latency benefit but
increases warehouse cost.