Anomali
Also known as: Anomali AURA, Anomali Unified Response Agent, Anomali Agentic SOC, Anomali ThreatStream
Anomali combines a security data lake, ThreatStream intelligence and an Agentic Fleet for security operations. AURA coordinates alert triage, incident investigation and case management. Agents correlate activity with threat and organizational context, route consequential decisions to an Approval Queue, and preserve evidence and outcomes for analysts. ThreatStream also gives external tools and buyer agents access to threat intelligence through APIs and MCP.
Anomali sells an Agentic SOC Platform for security teams that need to investigate alerts and maintain a defensible record of how incidents were handled. It ingests security data from endpoint, cloud, network, identity and other systems into its Unified Security Data Lake.
Its Intelligent Unification Layer normalizes and deduplicates events, then combines them with ThreatStream intelligence and context about the organization's assets and prior activity. The Decision Intelligence Graph makes that information searchable for analysts and other agents.
ThreatStream Next Gen can enrich telemetry as it arrives with curated intelligence about threat actors, campaigns and infrastructure, and Trusted Circles shares intelligence among ISAC and ISAO members.
Anomali Unified Response Agent, or AURA, coordinates a fleet with specialized roles. When automatic triage is enabled on an alert rule, the Alert Triage Agent evaluates generated alerts against account activity, entity risk and related evidence. The Incident and Investigation Agent can gather additional context after a handoff, and the Case Management Agent organizes findings and decisions into a case.
Long term memory about users and operations is retrieved at run time when it bears on the task, stays inside one organization and is not used to train the underlying model. For decisions such as promoting an alert to an incident, the Approval Queue shows the proposed action, confidence and supporting evidence. When a promotion is approved, AURA creates the incident and preserves the linked alert. Analysts retain responsibility for performing recommended response actions.
Administrators can pause the fleet or a particular agent. Manual requests run within the initiating user's access, while background work uses a designated service account when configured. Agent run details show the trigger, available context, tools invoked and output, and actions are recorded in an immutable audit history. Anomali's Security Operations Platform has a FedRAMP Rev5 Class C Moderate certification.
Anomali sells the platform directly and on AWS Marketplace. A listed configuration costs $520,000 for 12 months for 3,500 employees, 0.5 TB of daily data and six months of storage. ThreatStream and Copilot also have separate marketplace contract options, and STAXX is a free STIX and TAXII client for collecting threat feeds.
Anomali works alongside the SIEM and other security tools an organization already uses. Security teams can start by enriching their telemetry and linking repeated alerts to an investigation before expanding which workflows the agents handle.
Anomali serves more than 400 enterprise customers across financial services, critical infrastructure, government and national defense, including Admiral, Air Canada, the College Board and MISO Energy, and it took three Global InfoSec Awards at RSAC 2025, for SIEM, security operations and threat intelligence. One private charter airline consolidated its threat feeds with ThreatStream Next Gen and reduced investigation time from days to seconds. Other teams can use ThreatStream over REST or MCP to bring the same threat intelligence into existing defensive workflows.
Vendor details
Canonical URL
https://www.anomali.com/
Category
Security / SOC agent
Subcategory
Defensive alert triage and incident investigation
Funding status
Anomali is backed by Google Ventures, General Catalyst and IVP.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Security telemetry from SIEM, endpoint, cloud and identity systems. ThreatStream REST APIs and MCP, plus a Decision Intelligence Graph API for buyer agents.
In practice
An analyst receives related alerts and reviews AURA's evidence before approving a formal incident.
An intelligence team enriches alerts with ThreatStream and links the resulting context to an existing investigation.
An operations lead audits the tools and permissions used in one agent run and checks the case history for the final approval.
Sources & related URLs
Research sources
Agentic Index coverage score
9.5 / 14 capabilities · 68%
| Integrations & Tool Calling | Partial |
|---|---|
|
AURA can gather evidence from connected security telemetry and external threat intelligence. Enrichment draws on sources available to the organization and preserves that context as an alert moves to an investigation. The agents can create or update incidents and cases in Anomali. The current workflow leaves the recommended actions in outside security systems with the analyst. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/aura_overview.htmread 2026-11-20 |
|
| Workflow Orchestration | Full |
|
AURA coordinates the Alert Triage, Incident and Investigation, and Case Management agents as work progresses through a security case. The investigation agent can take over from triage with the relevant entities and rule context already attached. AURA tracks their work and passes the resulting decision or report back to the workflow that initiated it. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/aura_overview.htmread 2026-11-20 |
|
| Knowledge Grounding & RAG | Full |
|
The Decision Intelligence Graph keeps normalized telemetry searchable alongside curated threat intelligence, asset and identity context, and incident records. An investigation can query related security activity across sources while keeping its original alert as context. ThreatStream also enriches incoming data so fresh events are available to the same retrieval layer. Sourceanomali.comread 2026-11-20 |
|
| Human Oversight & Guardrails | Full |
|
The Approval Queue routes consequential recommendations to authorized people with evidence and confidence attached. A reviewer can approve, reject or override outcomes such as promoting an alert, escalating an incident or closing it. For an approved promotion, the platform can create the incident and retain the links back to the underlying alert automatically. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/approval_queue.htmread 2026-11-20 |
|
| Security, Identity & Governance | Full |
|
The Security Operations Platform is FedRAMP certified at Rev5 Class C Moderate. In the fleet, manually initiated tasks inherit the user's permissions and background tasks use assigned service account permissions. Security case access is configured separately, and agent fleet controls such as pausing all agents are restricted to Organization Administrators. Sourcefedramp.gov/marketplace/products/fr2527234065read 2026-11-20 |
|
| Observability & Auditability | Full |
|
My Agents links individual runs back to the event or request that triggered them. An analyst can inspect execution time, tools called, available evidence, permissions and the resulting decision. The audit trail distinguishes an agent action from a user or service action. It also records approvals, overrides and rejections for the security case history. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/agent_details.htmread 2026-11-20 |
|
| Memory & State Persistence | Partial |
|
AURA draws on long term user memory and operational memory at run time when they bear on the task, alongside prior agent work and earlier security activity. One organization's memory is never available to another, and none of it trains the underlying model. The fleet can also retain progress when a run is paused. The memory has no stated retention period or per user deletion path. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/aura_overview.htmread 2026-11-20 |
|
| Deployment & Data Residency | Not documented |
|
Anomali delivers the Security Operations Platform as a hosted service. Customers can connect logs and other sources from their own infrastructure, including from hybrid estates. The platform then normalizes and investigates those inputs in its hosted environment. Backend source location does not control where the Agentic Fleet or cloud data lake runs. Sourceaws.amazon.com/marketplace/pp/prodview-unjq3gvyofltiread 2026-11-20 |
|
| Prebuilt Agents / Templates / Packs | Full |
|
The platform includes purpose built Alert Triage, Incident and Investigation, and Case Management agents. They have separate tasks, individual run histories and controls in the Agentic Fleet overview. A team can start with the agent for a targeted part of its security workflow and monitor its activity before assigning more work to the wider Anomali fleet. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/fleet_overview.htmread 2026-11-20 |
|
| Triggers & Channel Coverage | Full |
|
An alert rule can invoke the Alert Triage Agent when new activity matches and automatic triage is enabled. Incoming detections and security events flow through configured rules before reaching the fleet. Analysts can also request investigation manually. The event driven path begins triage without requiring an analyst to open and select the new alert first. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/alert_triage_agent.htmread 2026-11-20 |
|
| Model Flexibility & Routing | Not documented |
|
A run can show the model used and the cost of the execution. The platform computes confidence from the available security evidence and uses its language model to explain that result. Users configure detection logic, permissions and approvals. The model displayed on completed investigation runs is a record of execution, with model selection handled by Anomali. Sourcedocs.anomali.com/content/nextgen/anomali_agentic_fleet/agent_details.htmread 2026-11-20 |
|
| APIs / SDKs / MCP Extensibility | Full |
|
ThreatStream Enterprise offers a REST API that supports retrieving intelligence, importing observables and managing investigations or threat entities. It uses scoped customer access with a user ID and API key. ThreatStream also makes intelligence available to outside agents through MCP, and the platform exposes its Decision Intelligence Graph for customer tools. Sourcedocs.anomali.com/content/threatstream%20api/introduction_api.htmread 2026-11-20 |
|
| Testing, Debugging & Optimization | Partial |
|
Detection rules pass two validation tiers before a team activates them. A structure check runs while the analyst edits, then a simulation against recent data checks volume, entity resolution and duplicate outcomes. That tests the detection logic that feeds the agents, not the agents' own triage and investigation calls. For the fleet itself, run details show the trigger, evidence, tools called and decision, so an analyst can inspect a run after the fact, with no scored test set or accuracy check on agent decisions. Sourcedocs.anomali.com/content/nextgen/security_ops/detection_rules/detection_rules_create.htmread 2026-11-20 |
|
| Browser / Computer-use | Not documented |
|
The fleet works over security telemetry, normalized records, threat intelligence and the platform's incident and case objects. Analysts inspect those results in the Anomali application and can follow a recommendation into a response workflow. Agent work in this service queries security data and changes case objects without controlling a browser or desktop session. Sourcedocs.anomali.com/content/nextgen/security_ops/secops_overview.htmread 2026-11-20 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
$520,000 per year for the listed Agentic SOC configuration
Annual contract with specified employee, ingest and storage capacity
Included quota
Listed platform configuration covers 3,500 employees, 0.5 TB per day and six months of storage.
What is public
Anomali offers a $520,000 annual platform configuration for 3,500 employees, 0.5 TB of daily data and six months of storage on AWS Marketplace. Its respective listings are $150,000 for ThreatStream Enterprise, $83,333 for Copilot Essential and $338,461 for ThreatStream AI Enterprise with 50 GB per day of indicator ingest.
Billing mechanics
Each marketplace dimension includes entitlement to the selected option for the 12 month contract. Platform capacity includes an employee count, a daily data volume and a retention window.
Variable cost rationale
Annual fees depend on capacity and package. Pricing varies with daily data volume, storage and purchased threat intelligence or Copilot options.
Sales call required
Yes, required for paid access
Commercial notes
Anomali sells the annual configuration through AWS Marketplace. It also sells the platform directly.
Missing data
Prices for configurations outside the listed dimensions are quoted.
Related vendors
- 7AI — Agentic SOC from the Cybereason founders: domain specialized agents…
- Abnormal AI — Behavioral AI email security with three named autonomous agents: a…
- AirMDR — AI native MDR whose AI analyst investigates every alert and writes…
- Airrived — Agentic OS that unifies SOC, GRC, IAM, vulnerability management, IT,…
- Andesite — Human-AI SOC platform where customers build their own agents and…
- Anvilogic — Agentic security operations platform that works on top of existing…
Alternatives to Anomali
Other products that do the work Anomali does, and what each one is.
- BlinkOpsAgentic security operations platform
- Command ZeroAutonomous and AI assisted SOC platform
- DrataTrust management platform
- Magnitude
- Splunk
- Sumo Logic