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
chore(openapi): regenerate spec for PR-5 (investigation queue endpoint)
Beenu Arora6438d4b1cce6
diff --git a/docs/openapi.yaml b/docs/openapi.yamlindex df9ce9b85..155600c09 100644--- a/docs/openapi.yaml+++ b/docs/openapi.yaml@@ -353,6 +353,83 @@ paths:$ref: '#/components/schemas/AlertStatsResponse'security:- HTTPBearer: []+ /api/v1/alerts/queue:+ get:+ tags:+ - alerts+ summary: Get Alert Queue+ description: 'Investigation Queue workbench feed.+++ Returns the prioritised list of alerts the analyst should work on++ next: alerts assigned to them, followed by unassigned++ critical/high alerts, ordered by SLA due time.+++ The endpoint also returns ``counts`` for both buckets unconditionally++ so the topbar badge and the workbench tabs can render without an++ extra round-trip.'+ operationId: get_alert_queue_api_v1_alerts_queue_get+ security:+ - HTTPBearer: []+ parameters:+ - name: owner+ in: query+ required: false+ schema:+ enum:+ - me+ - unassigned+ - all+ type: string+ default: all+ title: Owner+ - name: period+ in: query+ required: false+ schema:+ enum:+ - 24h+ - 7d+ - 30d+ - all+ type: string+ default: all+ title: Period+ - name: page+ in: query+ required: false+ schema:+ type: integer+ minimum: 1+ default: 1+ title: Page+ - name: page_size+ in: query+ required: false+ schema:+ type: integer+ maximum: 200+ minimum: 1+ default: 50+ title: Page Size+ responses:+ '200':+ description: Successful Response+ content:+ application/json:+ schema:+ $ref: '#/components/schemas/QueueResponse'+ '422':+ description: Validation Error+ content:+ application/json:+ schema:+ $ref: '#/components/schemas/HTTPValidationError'/api/v1/alerts/{alert_id}:get:tags:@@ -448,6 +525,43 @@ paths:application/json:schema:$ref: '#/components/schemas/HTTPValidationError'+ /api/v1/alerts/{alert_id}/claim:+ post:+ tags:+ - alerts+ summary: Claim Alert Endpoint+ description: 'Atomically claim an unassigned alert for the current user.+++ Returns ``409 Conflict`` if the alert is already assigned to someone++ else — the claim is a compare-and-set on ``assigned_to_id``, so two++ analysts racing for the same alert never both win.'+ operationId: claim_alert_endpoint_api_v1_alerts__alert_id__claim_post+ security:+ - HTTPBearer: []+ parameters:+ - name: alert_id+ in: path+ required: true+ schema:+ type: string+ format: uuid+ title: Alert Id+ responses:+ '200':+ description: Successful Response+ content:+ application/json:+ schema:+ $ref: '#/components/schemas/AlertResponse'+ '422':+ description: Validation Error+ content:+ application/json:+ schema:+ $ref: '#/components/schemas/HTTPValidationError'/api/v1/alerts/{alert_id}/snooze:post:tags:@@ -19683,6 +19797,224 @@ components:- rows- total_rowstitle: QueryResult+ QueueAction:+ properties:+ priority:+ type: integer+ title: Priority+ action:+ type: string+ title: Action+ risk:+ type: string+ title: Risk+ default: low+ type: object+ required:+ - priority+ - action+ title: QueueAction+ description: 'Top suggested next action for a queue row.+++ Sourced from ``Alert.ai_recommendations``. The lowest-priority++ integer wins (1 = most urgent). String-only legacy values are++ accepted and mapped to ``risk="low"`` with the array index as a++ fallback priority.'+ QueueAsset:+ properties:+ kind:+ type: string+ title: Kind+ description: host | user | ip | asset+ value:+ type: string+ title: Value+ label:+ anyOf:+ - type: string+ - type: 'null'+ title: Label+ type: object+ required:+ - kind+ - value+ title: QueueAsset+ description: 'Lightweight asset descriptor for a queue row.+++ The queue surfaces *one* representative entity per alert so the++ row is informative at a glance ("VPN brute force on `web-01`")++ without enumerating every entity. The full pivotable list lives++ on the Investigation Rail when the analyst opens the alert.'+ QueueItem:+ properties:+ id:+ type: string+ format: uuid+ title: Id+ tenant_id:+ type: string+ format: uuid+ title: Tenant Id+ title:+ type: string+ title: Title+ severity:+ type: string+ title: Severity+ status:+ type: string+ title: Status+ priority:+ type: integer+ title: Priority+ category:+ anyOf:+ - type: string+ - type: 'null'+ title: Category+ connector_type:+ anyOf:+ - type: string+ - type: 'null'+ title: Connector Type+ assigned_to_id:+ anyOf:+ - type: string+ format: uuid+ - type: 'null'+ title: Assigned To Id+ case_id:+ anyOf:+ - type: string+ format: uuid+ - type: 'null'+ title: Case Id+ first_seen:+ type: string+ format: date-time+ title: First Seen+ sla_due_at:+ type: string+ format: date-time+ title: Sla Due At+ sla_remaining_seconds:+ type: integer+ title: Sla Remaining Seconds+ sla_breached:+ type: boolean+ title: Sla Breached+ age_seconds:+ type: integer+ title: Age Seconds+ asset:+ anyOf:+ - $ref: '#/components/schemas/QueueAsset'+ - type: 'null'+ suggested_action:+ anyOf:+ - $ref: '#/components/schemas/QueueAction'+ - type: 'null'+ bucket:+ type: string+ enum:+ - mine+ - unassigned+ title: Bucket+ type: object+ required:+ - id+ - tenant_id+ - title+ - severity+ - status+ - priority+ - first_seen+ - sla_due_at+ - sla_remaining_seconds+ - sla_breached+ - age_seconds+ - bucket+ title: QueueItem+ description: 'One row in the Investigation Queue workbench.+++ Intentionally slimmer than ``AlertResponse``. The queue exists to++ answer "what should I work on next?" — clicking a row navigates++ to ``/alerts/{id}`` which loads full detail via the existing++ detail endpoint (``GET /alerts/{id}``). Keeping this payload++ small means a busy SOC with thousands of open alerts can refresh++ the queue cheaply (the badge on the topbar polls it).'+ QueueResponse:+ properties:+ items:+ items:+ $ref: '#/components/schemas/QueueItem'+ type: array+ title: Items+ total:+ type: integer+ title: Total+ counts:+ additionalProperties:+ type: integer+ type: object+ title: Counts+ period:+ type: string+ title: Period+ owner:+ type: string+ title: Owner+ page:+ type: integer+ title: Page+ page_size:+ type: integer+ title: Page Size+ pages:+ type: integer+ title: Pages+ generated_at:+ type: string+ format: date-time+ title: Generated At+ type: object+ required:+ - items+ - total+ - counts+ - period+ - owner+ - page+ - page_size+ - pages+ - generated_at+ title: QueueResponse+ description: 'Paginated queue response.+++ ``counts`` always reports both buckets so the workbench can++ render the "mine / unassigned / all" tabs without re-fetching.++ ``generated_at`` is the server-side ``now`` we used to compute++ SLA remaining — the frontend can drift its own clock from this++ so multiple analysts on different boxes see the same countdowns.'RatingIn:properties:score: