Proofpoint
Also known as: proofpoint.com, Proofpoint Satori, Satori Abuse Mailbox Agent, Satori DLP Triage Agent, Satori MCP Access, Proofpoint Secure Agent Gateway, Proofpoint Nexus
Proofpoint sells email, data loss and human risk security, and Satori is its platform for security operations agents. The Abuse Mailbox Agent classifies emails users report and decides the next step, the DLP Triage Agent clears false positives from DLP alerts and escalates the rest, and a third agent answers exploited vulnerability questions inside Microsoft Security Copilot. Satori MCP Access lets other vendors' agents call Satori, and separate products discover and police customers' own AI agents and MCP connections.
Proofpoint sells security for people and data: email protection against phishing, business email compromise and malware, email fraud defense, enterprise and adaptive email DLP, security awareness training and a growing AI security line, backed by the Proofpoint Nexus threat intelligence and AI layer. More than 80 of the Fortune 100 use it, it protects 3.4 million customers, and Accenture, Informatica, Wyndham and Concentrix are among them. Thoma Bravo took the company private in 2021, it bought Hornetsecurity in 2025 to reach managed service providers, and in October 2026 it agreed to buy Varonis for about $6.3 billion, a deal expected to close in the first half of 2027.
Proofpoint Satori is the agent platform for security operations, built to automate multistep security work within guardrails and under human supervision. The Satori Abuse Mailbox Agent reads the emails employees report, classifies each as phishing, malware, spam or clean, decides the next step and works inside Proofpoint CLEAR and the Threat Protection Workbench, where customer built workflows quarantine or tag messages on its verdict.
The Satori DLP Triage Agent reviews DLP alerts from endpoint, cloud and email sources in the Data Security Workbench, filters false positives and escalates incidents that need an analyst. A third agent, the Satori Emerging Threats Intelligence Agent, answers questions inside Microsoft Security Copilot about which vulnerabilities attackers are exploiting, and Satori MCP Access lets other agents such as CrowdStrike Charlotte and Microsoft Copilot call Satori over the Model Context Protocol.
Analysts can ask Satori in chat why it decided something, shape future handling with Satori Memories written in plain language, and review audit trails of its classifications.
For customers building their own agents, Secure AI Usage by Agents discovers custom and managed agents with their toolchains and MCP connections, watches multistep runs for behavioral drift, rebuilds each action from request to outcome and flags actions that go beyond the user's request, while Secure Agent Gateway, built on MCP, controls which agents reach which data and blocks or redacts sensitive data before it is shared.
Pricing is quote only through a demo request, and the Emerging Threats Intelligence Agent costs nothing extra for Security Copilot customers holding an Emerging Threats Intelligence license. Proofpoint holds ISO 27001 certification, and its Data Security console signs users in through SAML or OAuth identity providers and grants access through predefined or custom policies. Satori's processing runs on AWS in the US and on Azure in the US, Germany, Australia, Canada, Japan, Sweden, Singapore, the UAE and India, and several Proofpoint services place data by the customer's geography. Customers cannot pick or swap the model behind Satori.
It fits large Proofpoint email and DLP customers whose analysts drown in user reported phishing and DLP alerts, SOC teams that want their main assistant to hand email cases to a specialist agent, and security teams that need to see and govern the agents their own developers deploy. It fits less well for teams outside the Proofpoint stack, buyers who want agents acting across third party tools, or anyone needing a list price.
Vendor details
Canonical URL
https://www.proofpoint.com
Category
Security / SOC agent
Subcategory
Email and data security platform with Satori agents that triage reported email and DLP alerts, plus security for customers' AI agents and MCP connections
Funding status
Private, owned by Thoma Bravo since its take private closed in August 2021 in a deal valued at about $12.3 billion.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Satori works inside Proofpoint CLEAR, the Threat Protection Workbench and the Data Security Workbench, with endpoint, cloud and email DLP sources feeding it and customer built workflows acting on its verdicts. Satori MCP Access lets agents such as CrowdStrike Charlotte and Microsoft Copilot call Satori agents, and the Emerging Threats Intelligence Agent runs in Microsoft Security Copilot. The AI security products discover custom and managed agents, their toolchains and MCP connections, and Proofpoint integrates with the Claude Compliance API to govern Claude activity.
In practice
A bank's employees report 3,000 suspicious emails a week. The Satori Abuse Mailbox Agent classifies each one, the bank's CLEAR workflow quarantines the message whenever Satori calls it phishing, and analysts only review the handful it escalates.
A healthcare company's DLP tools fire hundreds of alerts a day on patient data. The Satori DLP Triage Agent clears the false positives, escalates the real exposures to the Data Security Workbench, and an analyst adds a Satori Memory telling it to treat a known billing partner's domain as approved.
A software company uses Secure AI Usage by Agents to find every agent its engineers deployed, then sees an agent that was asked to summarize a ticket trying to export a customer list; intent based access control flags the action and the runtime guardrail stops it.
Sources & related URLs
Related / legacy domains
Research sources
Agentic Index coverage score
8.0 / 14 capabilities · 57%
| Integrations & Tool Calling | Partial |
|---|---|
|
Satori's agents work inside Proofpoint's own consoles: the Abuse Mailbox Agent in Proofpoint CLEAR and the Threat Protection Workbench, the DLP Triage Agent in the Data Security Workbench, reading alerts that endpoint, cloud and email DLP sources feed in. Follow up steps such as quarantining or tagging a message run as workflows the customer builds on Satori's verdict. Integration with other vendors runs inward: Satori MCP Access lets agents such as CrowdStrike Charlotte and Microsoft Copilot call Satori, and the Emerging Threats Intelligence Agent answers inside Microsoft Security Copilot, while the agents themselves take no action in third party systems. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Workflow Orchestration | Partial |
|
Each Satori agent runs its own multistep job within guardrails: analyze the item, classify it, decide what should happen next and escalate when a person is needed. The steps after that live in customer created workflows in CLEAR and the Data Security Workbench, and Satori offers no builder for chaining several agents or designing a flow of its own. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Knowledge Grounding & RAG | Partial |
|
Satori grounds its verdicts in the message or alert in front of it, the context in Proofpoint's workbenches and the Proofpoint Nexus threat intelligence and AI layer behind the platform. It does not keep a searchable store of the customer's own documents or policies that it retrieves from when it decides. Analysts steer it instead through Satori Memories, which tell the agents how to treat known senders, domains or alert types. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Human Oversight & Guardrails | Partial |
|
Satori runs under human in the loop supervision: agents take only permitted actions within defined guardrails, escalate cases that need a person, and answer analysts' questions about their decisions in chat. Satori has no queue where each proposed action waits for an analyst's approval, since follow up actions run through workflows the customer sets. For customers' own agents, real time guardrails stop unsafe actions before they run and Secure Agent Gateway blocks or redacts sensitive data before an agent shares it. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Security, Identity & Governance | Full |
|
Proofpoint holds ISO 27001 certification for its information security management system, with technical and organizational measures set in its data processing agreement. The Data Security console where the DLP Triage Agent works grants access through policies that bundle read and write capabilities, from the predefined Full Administration, Activity View and List View policies to custom ones, assigned to users or groups. Sign in runs through SAML or OAuth 2.0 identity providers such as Okta, Microsoft Entra ID, Google Workspace and OneLogin, and only portal users with the right permissions can create tokens for the Proofpoint API. Sourcedocs.public.analyze.proofpoint.com/product_overview/getting_started_with_the_platform.htmread 2026-11-18 |
|
| Observability & Auditability | Full |
|
Audit trails record every Satori classification alongside the analysts' interactions with it, for review and accountability, and analysts can question the agent about any decision. For the customer's own agents, AI Agent Runtime Observability follows multistep and multi agent runs for behavioral drift and forensics rebuild each action from the original request to its outcome. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Memory & State Persistence | Partial |
|
Satori Memories let analysts write instructions in plain language that shape how the agents handle similar items from then on, such as treating a known partner's domain as approved or routing a recurring alert type straight to a named team. The store is written by people rather than learned by the agent from its own runs, and it carries policy guidance rather than a record of each past case. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Deployment & Data Residency | Full |
|
Satori processes data on AWS in the US and on Azure in the US, Germany, Australia, Canada, Japan, Sweden, Singapore, the UAE and India, and for many of Proofpoint's services the data center is set by the customer's geographic location. Email Protection runs in a similar spread across North America, Europe, the Middle East and Asia Pacific. Placement follows where the customer is rather than a region picked from a list. Sourceproofpoint.com/us/legal/trust/subprocessorsread 2026-11-18 |
|
| Prebuilt Agents / Templates / Packs | Full |
|
Three ready agents ship under the Satori name, each a whole product for its own job: the Abuse Mailbox Agent for phishing reports, the DLP Triage Agent for data loss alerts, and the Emerging Threats Intelligence Agent, generally available in Microsoft Security Copilot since November 2025 at no extra cost for Emerging Threats Intelligence license holders, which tells defenders which vulnerabilities are being exploited. Customers switch them on rather than build them, and shape the first two afterward with memories. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Triggers & Channel Coverage | Full |
|
Work arrives without an analyst starting it. Every email an employee reports to the abuse mailbox goes to the Abuse Mailbox Agent, and every DLP alert from endpoint, cloud and email sources goes to the DLP Triage Agent, which analyze and classify them autonomously as they come in. The AI security products likewise monitor customers' agents and MCP connections continuously at runtime. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Model Flexibility & Routing | Not documented |
|
Satori runs on models Proofpoint chooses and keeps unnamed, and customers have no setting to choose, route or swap a model for any of its agents. Proofpoint Nexus is the company's own AI and threat intelligence layer behind the platform rather than a choice of model, so the same engine handles every verdict whatever the customer's preference or data rules. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| APIs / SDKs / MCP Extensibility | Partial |
|
Developers reach Proofpoint through the Proofpoint API, whose tokens permissioned portal users create and revoke under Integrations, and through the Targeted Attack Protection API that feeds threat events and clicks into outside tools. Satori MCP Access opens Satori agents to other vendors' agents over the Model Context Protocol, with CrowdStrike Charlotte and Microsoft Copilot named as callers, so a SOC assistant elsewhere can hand a reported email or an alert to Satori. The calling surface invokes agents rather than exposing a full set of tools over Proofpoint's own policies and records. Sourceproofpoint.com/us/newsroom/press-releases/proofpoint-secures-collaboration-and-data-agentic-workspace-industry-firstread 2026-11-18 |
|
| Testing, Debugging & Optimization | Not documented |
|
Satori's accuracy is judged by analysts reviewing its verdicts and adding memories where it went wrong. There are no test sets, scored evaluations or red team runs against Satori or against the customer's own agents, whose behavior the AI security products watch at runtime instead, and no way to compare one version of Satori's handling against another before it goes live, so its error rate comes from sampling verdicts by hand. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
| Browser / Computer-use | Not documented |
|
The agents read reported messages and alerts inside Proofpoint's consoles and pass results to workflows. None of them drives a browser, opens suspicious links in a session it controls or operates another application's interface on an analyst's behalf. Link and attachment analysis belongs to Proofpoint's threat protection services, such as URL Defense and Attachment Defense, rather than to an agent working a screen. Sourceproofpoint.com/us/platform/satoriread 2026-11-18 |
|
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
Quote only; sold through a demo request
Enterprise subscriptions quoted per product
What is public
Proofpoint sells Satori and its AI security products by quote, after a demo. The Satori Emerging Threats Intelligence Agent costs nothing extra for Microsoft Security Copilot customers who hold a Proofpoint Emerging Threats Intelligence license. Proofpoint estimates that Satori frees half of analyst workload and returns more than 500% when triaging 10,000 alerts.
Billing mechanics
Each product is quoted on its own, and the Satori agents attach to the email and data security products whose alerts they triage.
Cost watchouts
Satori's agents work inside Proofpoint CLEAR, the Threat Protection Workbench and the Data Security Workbench, so buyers need those products in place for the agents to act on, and the Security Copilot agent needs both a Security Copilot subscription and an Emerging Threats Intelligence license.
Variable cost rationale
Contracts are quoted subscriptions rather than metered agent usage.
Additional watchouts
No list prices appear for any product.
Sales call required
Yes, required for paid access
Free / trial
No free plan or trial; demos on request
Lowest paid plan
None; Satori and the AI security products are quoted
Commercial notes
The return and workload figures are Proofpoint's own estimates.
Missing data
Proofpoint gives no prices for Satori, its agents or the AI security products.
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 Proofpoint
Other products that do the work Proofpoint does, and what each one is.
- Horizon3.aiNodeZero autonomous penetration testing across internal
- NetskopeCloud security platform
- Operant AIRuntime security for AI agents, MCP servers and employee AI tools
- Qevlar AI
- IdiraIdentity security platform extended to AI agents
- Mimecast