# Security Operations Center

A Security Operations Center (SOC) is a team or centralized function that continuously monitors, detects, investigates, and responds to cybersecurity threats and incidents.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## It's time to retire vanity security dashboards

DevFeed: [It's time to retire vanity security dashboards](<https://devfeed.tech/articles/it-s-time-to-retire-vanity-security-dashboards-145726.md>)

Original publisher: [Read original article](<https://www.helpnetsecurity.com/2026/11/20/security-dashboards-tips-video/>)

Author: Help Net Security

Published: 2026-11-20T06:00:21Z

Content type: opinion

Language: en

Sources: [Help Net Security](<https://devfeed.tech/sources/help-net-security.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Data visualization](<https://devfeed.tech/topics/data-visualization.md>)

Tags: [alert](<https://devfeed.tech/tags/alert.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [control](<https://devfeed.tech/tags/control.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [dashboards](<https://devfeed.tech/tags/dashboards.md>), [don-t-miss](<https://devfeed.tech/tags/don-t-miss.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [environments](<https://devfeed.tech/tags/environments.md>), [it](<https://devfeed.tech/tags/it.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [news](<https://devfeed.tech/tags/news.md>), [ot](<https://devfeed.tech/tags/ot.md>), [ot-security](<https://devfeed.tech/tags/ot-security.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [review](<https://devfeed.tech/tags/review.md>), [security](<https://devfeed.tech/tags/security.md>), [soc](<https://devfeed.tech/tags/soc.md>), [strategy](<https://devfeed.tech/tags/strategy.md>), [teams](<https://devfeed.tech/tags/teams.md>), [tickets](<https://devfeed.tech/tags/tickets.md>), [tips](<https://devfeed.tech/tags/tips.md>), [triggers](<https://devfeed.tech/tags/triggers.md>), [video](<https://devfeed.tech/tags/video.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

Mahamed argues that security dashboards should measure reduced exposure and effective follow-up, not ticket closure counts. At a manufacturing site, recurring alerts from the same approved maintenance work led analysts to close thousands of tickets without resolving the underlying issue. He recommends that SOC and control engineering teams review such patterns together, add operational context when tuning detection rules, and verify that unexpected activity still triggers alerts. For vulnerability dashboards, he notes that a lower-severity issue on a server reachable through a supplier connection may warrant more attention than a high-severity issue on a restricted workstation. Dashboards should show what exposure was removed, who owns the next action, and whether it worked.

### Source excerpt

In this Help Net Security video, Abdul Khadir Mahamed, OT Security Architect at Estée Lauder Companies, explains why security teams should retire vanity dashboards in IT and OT environments. He walks through an example from a manufacturing site, where the same approved maintenance work triggers the same alert week after week. Analysts close thousands of tickets, but the underlying problem stays. He describes how SOC and control engineering teams can review these patterns together, tune ... More -> The post It's time to retire vanity security dashboards appeared first on Help Net Security.

## RMM Alerts at 2 AM: An MSP and SOC Incident Response Story

DevFeed: [RMM Alerts at 2 AM: An MSP and SOC Incident Response Story](<https://devfeed.tech/articles/rmm-alerts-at-2-am-an-msp-and-soc-incident-response-story-146283.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/rmm-alerts-msp-soc-incident-response>)

Author: Beth Robinson

Published: 2026-11-19T08:00:00Z

Content type: news

Language: en

Sources: [Huntress Blog](<https://devfeed.tech/sources/huntress-blog.md>)

Topics: [Incident response](<https://devfeed.tech/topics/incident-response.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Endpoint security](<https://devfeed.tech/topics/endpoint-security.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>)

Tags: [alerts](<https://devfeed.tech/tags/alerts.md>), [incident](<https://devfeed.tech/tags/incident.md>), [incident-response](<https://devfeed.tech/tags/incident-response.md>), [msp](<https://devfeed.tech/tags/msp.md>), [rmm](<https://devfeed.tech/tags/rmm.md>), [screenconnect](<https://devfeed.tech/tags/screenconnect.md>), [soc](<https://devfeed.tech/tags/soc.md>)

### AI overview

A compromised account with broad admin rights was used to deploy rogue ScreenConnect instances across nearly 1,000 endpoints. Huntress's 24/7 SOC isolated affected endpoints, sent a critical report, and joined TruePath IT and the client to discuss response decisions. Under pressure to resume operations, some still-compromised machines returned to use; 12 days later, another rogue deployment appeared, apparently delivering ransomware. TruePath locked the environment down again and completed cleanup. Afterward, endpoints returned to production only after the client's IT team verified remediation on each one and an executive gave verbal approval. The excerpt recommends agreeing in advance who can isolate systems and make decisions, who communicates with leadership, and what downtime and recovery the business can tolerate. It also suggests defining approved remote access tools, monitoring normal software before blocking other tools, removing unneeded software, and reviewing internet-exposed ports and services.

### Source excerpt

One compromised account, two rogue ScreenConnect deployments, and $3M in downtime. See how TruePath IT and the Huntress SOC handled RMM alerts at 2am.

## Huntress Delivers 24/7 Response for Microsoft Defender Incidents Across Endpoints and Microsoft 365

DevFeed: [Huntress Delivers 24/7 Response for Microsoft Defender Incidents Across Endpoints and Microsoft 365](<https://devfeed.tech/articles/huntress-delivers-24-7-response-for-microsoft-defender-incidents-across-endpoints-and-microsoft-365-140705.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/huntress-response-for-microsoft-defender-incidents-across-endpoints-and-microsoft-365>)

Author: Aimee Simpson; Andrew Schwartz

Published: 2026-11-17T14:00:00Z

Content type: release

Language: en

Sources: [Huntress Blog](<https://devfeed.tech/sources/huntress-blog.md>)

Topics: [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [microsoft 365](<https://devfeed.tech/topics/microsoft-365.md>), [Endpoint Security & XDR](<https://devfeed.tech/topics/endpoint-security-xdr.md>), [Security, Identity, & Compliance](<https://devfeed.tech/topics/security-identity-compliance.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-security](<https://devfeed.tech/tags/agentic-security.md>), [apps](<https://devfeed.tech/tags/apps.md>), [campaign](<https://devfeed.tech/tags/campaign.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [data](<https://devfeed.tech/tags/data.md>), [defender](<https://devfeed.tech/tags/defender.md>), [entra-id](<https://devfeed.tech/tags/entra-id.md>), [huntress](<https://devfeed.tech/tags/huntress.md>), [identity](<https://devfeed.tech/tags/identity.md>), [incidents](<https://devfeed.tech/tags/incidents.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-365](<https://devfeed.tech/tags/microsoft-365.md>), [microsoft-defender](<https://devfeed.tech/tags/microsoft-defender.md>), [office](<https://devfeed.tech/tags/office.md>), [office-365](<https://devfeed.tech/tags/office-365.md>), [organizations](<https://devfeed.tech/tags/organizations.md>), [platform](<https://devfeed.tech/tags/platform.md>), [protection](<https://devfeed.tech/tags/protection.md>), [response](<https://devfeed.tech/tags/response.md>)

### AI overview

Huntress now ingests incidents from Defender for Identity, Defender for Office 365, Defender for Cloud Apps, and Entra ID Protection, adding to its existing Defender for Endpoint and Antivirus signals. Its SOC triages incidents around affected endpoints or identities, investigates Microsoft tenant data through advanced hunting for up to 30 days, and reports findings and actions; Huntress applies per-tenant query limits to avoid using the tenant's shared Microsoft quota. The article reports that combined Huntress and Microsoft signals helped investigate a token-replay campaign affecting 97 accounts across 79 organizations from July 14 to September 22, 2026. Attackers used stolen session tokens, then registered devices and authenticator apps and created inbox rules. For suspected token replay, it recommends revoking sessions and resetting the password, then auditing Entra ID for attacker-registered devices and MFA methods, since a password reset alone leaves them in place.

### Source excerpt

The Huntress Agentic Security Platform now sees incidents from Defender for Identity, Defender for Office 365, Defender for Cloud Apps, and Entra ID Protection. The new data already helped uncover a token-replay campaign that hit 97 accounts at 79 organizations.

## EduSOC: A Lightweight Simulated Security Operations Center with AI Incident Responders

DevFeed: [EduSOC: A Lightweight Simulated Security Operations Center with AI Incident Responders](<https://devfeed.tech/articles/edusoc-a-lightweight-simulated-security-operations-center-with-ai-incident-responders-139231.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.17703>)

Author: Martin Higgins, Shawn Thompson, Cherry Mangla, Bobby Swar

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-ai](<https://devfeed.tech/tags/artificial-intelligence-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [cybersecurity-education](<https://devfeed.tech/tags/cybersecurity-education.md>), [decision-making](<https://devfeed.tech/tags/decision-making.md>), [education](<https://devfeed.tech/tags/education.md>), [engagement](<https://devfeed.tech/tags/engagement.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [event](<https://devfeed.tech/tags/event.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [investigations](<https://devfeed.tech/tags/investigations.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [lightweight](<https://devfeed.tech/tags/lightweight.md>), [lightweight-ai](<https://devfeed.tech/tags/lightweight-ai.md>), [operations](<https://devfeed.tech/tags/operations.md>), [platform](<https://devfeed.tech/tags/platform.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [security-operations-center](<https://devfeed.tech/tags/security-operations-center.md>)

### AI overview

EduSOC is a lightweight web-based SOC simulation for cybersecurity education, designed to focus introductory exercises on investigative decisions without the deployment demands of conventional cyber ranges. It generates synthetic events continuously, provides searchable evidence, and uses role-based AI responders that support investigations without disclosing incident classifications. Students interpret the evidence, consult simulated operational teams, and justify escalation decisions. In a preliminary evaluation of three university cohorts with 50 matched pre- and post-simulation survey participants, students reported high engagement and improved SOC self-efficacy; short-term knowledge gains varied across cohorts.

### Source excerpt

arXiv:2605.17703v3 Announce Type: replace Abstract: This paper presents EduSOC, a lightweight web-based Security Operations Center (SOC) simulation platform for cybersecurity education. Conventional cyber ranges can be complex and resource-intensive to deploy, particularly for introductory exercises focused on investigative decision-making rather than infrastructure configuration. EduSOC provides an alternative by combining continuous synthetic event generation, searchable supporting evidence, and role-based artificial intelligence (AI) responders that assist investigations without revealing underlying incident classifications. Students remain responsible for interpreting evidence, consulting simulated operational teams, and justifying their escalation decisions. A preliminary classroom evaluation was conducted across three university cohorts, comprising 50 matched pre- and post-simulation survey participants. The findings indicated high student engagement and improvements in self-reported SOC self-efficacy, while short-term knowledge gains varied across cohorts. These findings provide empirical insight into the use of lightweight, AI-supported SOC simulations in university cybersecurity education.

## New Huntress Security Insights Report & Reporting Center

DevFeed: [New Huntress Security Insights Report & Reporting Center](<https://devfeed.tech/articles/new-huntress-security-insights-report-reporting-center-136865.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/security-insights-report>)

Author: Mandy Ruggiero

Published: 2026-11-16T15:00:00Z

Content type: release

Language: en

Sources: [Huntress Blog](<https://devfeed.tech/sources/huntress-blog.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Threat monitoring](<https://devfeed.tech/topics/threat-monitoring.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>), [Endpoint security](<https://devfeed.tech/topics/endpoint-security.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>), [Security Information and Event Management (SIEM)](<https://devfeed.tech/topics/siem-security.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [addition](<https://devfeed.tech/tags/addition.md>), [advanced](<https://devfeed.tech/tags/advanced.md>), [all](<https://devfeed.tech/tags/all.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [announcement](<https://devfeed.tech/tags/announcement.md>), [attack](<https://devfeed.tech/tags/attack.md>), [huntress](<https://devfeed.tech/tags/huntress.md>), [insights](<https://devfeed.tech/tags/insights.md>), [one](<https://devfeed.tech/tags/one.md>), [one-page-summary](<https://devfeed.tech/tags/one-page-summary.md>), [report](<https://devfeed.tech/tags/report.md>), [reporting](<https://devfeed.tech/tags/reporting.md>), [security](<https://devfeed.tech/tags/security.md>), [security-insights](<https://devfeed.tech/tags/security-insights.md>), [summary](<https://devfeed.tech/tags/summary.md>), [threat](<https://devfeed.tech/tags/threat.md>), [threat-report](<https://devfeed.tech/tags/threat-report.md>), [updated](<https://devfeed.tech/tags/updated.md>)

### AI overview

Huntress is rolling out redesigned monthly and quarterly Security Insights Reports, renamed from Threat Reports, to show security posture and outcomes across Managed EDR, ITDR, SIEM, and ISPM. The report adds an executive summary with a protection funnel, SOC notes, and incident trends; firewall status and external attack-surface pages; and a cross-product incident summary. A separate one-page report gives nontechnical stakeholders a plain-language view of protection and investigations; it is emailed monthly and available in the portal. Both the new and legacy reports will be provided for a few months during transition. The Investigations View supplies a chronological timeline that includes benign closures. Planned additions include account-level reports, more Reporting Center customization, and a dedicated ISPM compliance report.

### Source excerpt

Meet the redesigned Huntress Security Insights Report (formerly known as Threat Report), a one-page summary in an updated reporting center.

## Exaforce's ExaU lets security staff spin up a personal team of AI agents

DevFeed: [Exaforce's ExaU lets security staff spin up a personal team of AI agents](<https://devfeed.tech/articles/exaforce-s-exau-lets-security-staff-spin-up-a-personal-team-of-ai-agents-135282.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/16/exaforces-exau-lets-security-staff-spin-up-a-personal-team-of-ai-agents/>)

Author: Duncan Riley

Published: 2026-11-16T12:00:11Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [activity-audit](<https://devfeed.tech/tags/activity-audit.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-security](<https://devfeed.tech/tags/agentic-security.md>), [agentic-security-operations](<https://devfeed.tech/tags/agentic-security-operations.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [alert-triage](<https://devfeed.tech/tags/alert-triage.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-agents](<https://devfeed.tech/tags/artificial-intelligence-agents.md>), [audit-log](<https://devfeed.tech/tags/audit-log.md>), [chief-information-security-officer](<https://devfeed.tech/tags/chief-information-security-officer.md>), [exabots](<https://devfeed.tech/tags/exabots.md>), [exaforce](<https://devfeed.tech/tags/exaforce.md>), [exaforce-ai-security](<https://devfeed.tech/tags/exaforce-ai-security.md>), [exaforce-inc](<https://devfeed.tech/tags/exaforce-inc.md>), [exau](<https://devfeed.tech/tags/exau.md>), [first](<https://devfeed.tech/tags/first.md>), [funding](<https://devfeed.tech/tags/funding.md>), [harbourvest-partners](<https://devfeed.tech/tags/harbourvest-partners.md>), [identity-providers](<https://devfeed.tech/tags/identity-providers.md>), [inc](<https://devfeed.tech/tags/inc.md>), [khosla-ventures](<https://devfeed.tech/tags/khosla-ventures.md>), [knowledge-graph](<https://devfeed.tech/tags/knowledge-graph.md>), [lets](<https://devfeed.tech/tags/lets.md>), [mayfield-fund](<https://devfeed.tech/tags/mayfield-fund.md>), [news](<https://devfeed.tech/tags/news.md>), [peak-xv-partners](<https://devfeed.tech/tags/peak-xv-partners.md>), [penetration-testing](<https://devfeed.tech/tags/penetration-testing.md>), [personal](<https://devfeed.tech/tags/personal.md>), [risk-assessment](<https://devfeed.tech/tags/risk-assessment.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [security-telemetry](<https://devfeed.tech/tags/security-telemetry.md>), [seligman-ventures](<https://devfeed.tech/tags/seligman-ventures.md>), [series-b](<https://devfeed.tech/tags/series-b.md>), [spin](<https://devfeed.tech/tags/spin.md>), [staff](<https://devfeed.tech/tags/staff.md>), [team](<https://devfeed.tech/tags/team.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [threat-hunt](<https://devfeed.tech/tags/threat-hunt.md>), [up](<https://devfeed.tech/tags/up.md>)

### AI overview

ExaU lets security staff create personal AI agent teams by describing responsibilities in plain language or using templates. Agents retain decisions and instructions, determine next steps, and return discoveries, questions, and recommendations. They can coordinate work across teams--for example, combining a threat hunt, activity audit, and employee outreach into one assessment--and a CISO briefing agent can gather status across teams. ExaU runs on customer data within the Exaforce platform, which the article says already connects to identity providers, cloud services, and security telemetry; Exaforce claims its knowledge graph gives agents system context from the start. Investigations run autonomously, but actions affecting employees, systems, or communications require approval, and agent steps are recorded for later review. ExaU complements Exaforce-managed Exabots, which handle detection, triage, and investigations. It is available as part of the platform.

### Source excerpt

Agentic security operations startup Exaforce Inc. today announced the launch of ExaU, a service that lets any security staff member spin up a personal team of artificial intelligence agents. The new service seeks to move past the first wave of AI agents in security, which Exaforce says were built to offload a single task such [...] The post Exaforce's ExaU lets security staff spin up a personal team of AI agents appeared first on SiliconANGLE.

## Daylight Lands $80 Million for AI-Powered Security Operations

DevFeed: [Daylight Lands $80 Million for AI-Powered Security Operations](<https://devfeed.tech/articles/daylight-lands-80-million-for-ai-powered-security-operations-130304.md>)

Original publisher: [Read original article](<https://www.securityweek.com/daylight-lands-80-million-for-ai-powered-security-operations/>)

Author: Ionut Arghire

Published: 2026-11-12T13:12:11Z

Content type: news

Language: en

Sources: [SecurityWeek](<https://devfeed.tech/sources/securityweek.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Security](<https://devfeed.tech/topics/security.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-native](<https://devfeed.tech/tags/ai-native.md>), [apac](<https://devfeed.tech/tags/apac.md>), [aura](<https://devfeed.tech/tags/aura.md>), [cyber-threats](<https://devfeed.tech/tags/cyber-threats.md>), [cybersecurity-funding](<https://devfeed.tech/tags/cybersecurity-funding.md>), [daylight](<https://devfeed.tech/tags/daylight.md>), [emea](<https://devfeed.tech/tags/emea.md>), [funding](<https://devfeed.tech/tags/funding.md>), [product-development](<https://devfeed.tech/tags/product-development.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [us](<https://devfeed.tech/tags/us.md>)

### AI overview

Daylight raised $80 million in a Series B round, bringing its total funding to $120 million. Its Aura platform uses AI agents to hunt, analyze, and contain threats in real time, with human experts governing and improving operations. Customers can choose fully managed or co-managed service, or support from forward-deployed engineers. Aura also provides access to security data and security operations management, and covers organizational AI platforms against prompt injection, risky MCPs, and rogue AI-driven activity. Daylight plans to use the funding to develop the platform, expand its security and technology teams and US presence, and enter markets in EMEA and APAC.

### Source excerpt

The company plans to accelerate product development, hire new talent, expand its US presence, and enter new markets. The post Daylight Lands $80 Million for AI-Powered Security Operations appeared first on SecurityWeek.

## Embed Security lets analysts teach its AI SOC Platform how to investigate

DevFeed: [Embed Security lets analysts teach its AI SOC Platform how to investigate](<https://devfeed.tech/articles/embed-security-lets-analysts-teach-its-ai-soc-platform-how-to-investigate-129109.md>)

Original publisher: [Read original article](<https://www.helpnetsecurity.com/2026/11/11/embed-security-expands-ai-soc-platform/>)

Author: Industry News

Published: 2026-11-11T13:54:12Z

Content type: release

Language: en

Sources: [Help Net Security](<https://devfeed.tech/sources/help-net-security.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [knowledge-engineering](<https://devfeed.tech/topics/knowledge-engineering.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-soc](<https://devfeed.tech/tags/ai-soc.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [build](<https://devfeed.tech/tags/build.md>), [built-in](<https://devfeed.tech/tags/built-in.md>), [can](<https://devfeed.tech/tags/can.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [customer](<https://devfeed.tech/tags/customer.md>), [decision](<https://devfeed.tech/tags/decision.md>), [detection](<https://devfeed.tech/tags/detection.md>), [detection-engineering](<https://devfeed.tech/tags/detection-engineering.md>), [embed-security](<https://devfeed.tech/tags/embed-security.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [environment](<https://devfeed.tech/tags/environment.md>), [extend](<https://devfeed.tech/tags/extend.md>), [industry-news](<https://devfeed.tech/tags/industry-news.md>), [lets](<https://devfeed.tech/tags/lets.md>), [platform](<https://devfeed.tech/tags/platform.md>), [security](<https://devfeed.tech/tags/security.md>), [soc](<https://devfeed.tech/tags/soc.md>)

### AI overview

Embed Security announced three capabilities for adapting its AI SOC Platform to an organization's investigative practices. Customer-built iSteps let analysts describe investigative approaches in natural language; Embed turns them into autonomous investigations on its orchestration framework, alongside expert-built iSteps that combine deterministic logic and LLM reasoning. NoiseIQ Memories dynamically connects organizational context with investigation evidence, including context that changes over time. Detection Engineering analyzes investigated alerts across endpoint, cloud, email, and identity to show detection performance and recommend improvements to customer-authored detections in supported SIEM platforms. The excerpt describes these as product capabilities and recommendations; it does not name supported SIEM platforms or provide performance evidence.

### Source excerpt

Embed Security announced new capabilities that let security teams teach Embed how they investigate, what matters in their environment, and how to improve their detections. With customer-built iSteps, expanded NoiseIQ Memories, and Detection Engineering, organizations can extend Embed's built-in security expertise with their own knowledge and operating practices. Rather than programming workflows, analysts can teach Embed what they know, and its decision layer applies that expertise autonomously. "Everyone is telling security teams to build agents. ... More -> The post Embed Security lets analysts teach its AI SOC Platform how to investigate appeared first on Help Net Security.

## Model-Based Authoring of SOC Simulation Exercises for Cybersecurity Education

DevFeed: [Model-Based Authoring of SOC Simulation Exercises for Cybersecurity Education](<https://devfeed.tech/articles/model-based-authoring-of-soc-simulation-exercises-for-cybersecurity-education-128921.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14227>)

Author: Felix H\\"arer

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [test](<https://devfeed.tech/topics/test.md>)

Tags: [assessment](<https://devfeed.tech/tags/assessment.md>), [authoring](<https://devfeed.tech/tags/authoring.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [cybersecurity-education](<https://devfeed.tech/tags/cybersecurity-education.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [exercise](<https://devfeed.tech/tags/exercise.md>), [generative](<https://devfeed.tech/tags/generative.md>), [glm](<https://devfeed.tech/tags/glm.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [model](<https://devfeed.tech/tags/model.md>), [phishing](<https://devfeed.tech/tags/phishing.md>), [prototype](<https://devfeed.tech/tags/prototype.md>), [ransomware](<https://devfeed.tech/tags/ransomware.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [soc](<https://devfeed.tech/tags/soc.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

The paper presents a model-based method for building Virtual SOC exercises that keeps incident events and evidence aligned with alerts, learner tasks, timeline entries, and assessments. A phishing-to-ransomware prototype exercise combines eight events, eight objects, six evidence items, seven alerts, 13 timeline entries, and five assessment items. GLM-5.3 generates selected alert descriptions from the model; a separate validator checks references, ordering, evidence links, assessment completeness, and fixed fields. In the demonstrated validation, valid reference data produced true negatives and deliberately introduced inconsistencies produced true positives. The resulting package ran in the learner environment, supporting the feasibility of this approach in the tested exercise.

### Source excerpt

arXiv:2611.14227v1 Announce Type: new Abstract: Cybersecurity training and classroom exercises with a virtual Security Operations Center (SOC) combine alerts, evidence, timeline entries, learner tasks, and assessment items that must remain consistent across an evolving incident. This paper presents a model-based authoring method in which a technical and educational specification is used to construct an incident exercise model and a large language model (LLM) generates selected scenario text based on that model. The incident exercise model combines an incident model of events, objects, and evidence with an exercise orchestration model of SOC alerts, timeline entries, learner tasks, and assessment items. The method is implemented in a Virtual SOC prototype and evaluated using a phishing-to-ransomware reference exercise. Its incident exercise model contains eight incident events, eight incident objects, six evidence items, seven alerts in five batches, 13 timeline entries including 12 learner tasks, and five representative assessment items. GLM-5.3-generated alert descriptions are inserted and a separate validator checks references, ordering, evidence links, assessment completeness, and selected fixed fields. Validation is demonstrated with reference data, resulting in true-negative outcomes for valid data and true-positive outcomes for deliberately placed inconsistencies. The resulting exercise package is executable within the Virtual SOC learner environment. Overall, the evaluation indicates consistency across the involved model elements and establishes the feasibility of the model-based and generative approach.

## Sasha Roshan Takes Security Education Beyond the Demo

DevFeed: [Sasha Roshan Takes Security Education Beyond the Demo](<https://devfeed.tech/articles/sasha-roshan-takes-security-education-beyond-the-demo-124128.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/employee-spotlight-sasha-roshan>)

Author: Erin Bortz

Published: 2026-11-09T14:00:00Z

Content type: article

Language: en

Sources: [Huntress Blog](<https://devfeed.tech/sources/huntress-blog.md>)

Topics: [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [Security](<https://devfeed.tech/topics/security.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Incident response](<https://devfeed.tech/topics/incident-response.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Sports](<https://devfeed.tech/topics/sports.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [advice](<https://devfeed.tech/tags/advice.md>), [also](<https://devfeed.tech/tags/also.md>), [ambition](<https://devfeed.tech/tags/ambition.md>), [arts](<https://devfeed.tech/tags/arts.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attention](<https://devfeed.tech/tags/attention.md>), [avoid](<https://devfeed.tech/tags/avoid.md>), [based](<https://devfeed.tech/tags/based.md>), [bbc](<https://devfeed.tech/tags/bbc.md>), [best](<https://devfeed.tech/tags/best.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [demo](<https://devfeed.tech/tags/demo.md>), [difference](<https://devfeed.tech/tags/difference.md>), [education](<https://devfeed.tech/tags/education.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [huntress](<https://devfeed.tech/tags/huntress.md>), [sales](<https://devfeed.tech/tags/sales.md>), [security](<https://devfeed.tech/tags/security.md>), [security-education](<https://devfeed.tech/tags/security-education.md>)

### AI overview

Sasha Roshan's Sales Engineering work at Huntress combines technical customer support with cybersecurity education. In a partner conversation, he focused on overnight operations, false positives, decision-making, and how Huntress's SOC works alongside Microsoft Defender; that discussion led to a proposed trial and deployment-planning follow-up. Beyond one-to-one conversations, Roshan contributes to livestreams, workshops, videos, presentations, and Huntress Wire, aiming to make topics such as identity threats easier to understand. He says creating educational content and speaking with audiences also sharpen his explanations in customer work. He now helps lead Huntress Managed ISPM for the Sales Engineering team. His advice to aspiring Sales Engineers is to keep learning, ask good questions, and communicate clearly; the article notes that Huntress provides tools and resources.

### Source excerpt

Meet Sasha Roshan, Senior Sales Engineer at Huntress, a passionate advocate for cybersecurity. See how he makes a difference every day.

## How Greenlight Financial built an autonomous SOC solution on Amazon Bedrock AgentCore

DevFeed: [How Greenlight Financial built an autonomous SOC solution on Amazon Bedrock AgentCore](<https://devfeed.tech/articles/how-greenlight-financial-built-an-autonomous-soc-solution-on-amazon-bedrock-agentcore-121743.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/how-greenlight-financial-built-an-autonomous-soc-solution-on-amazon-bedrock-agentcore/>)

Author: Adam Maksimuk

Published: 2026-11-06T16:55:05Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Amazon Bedrock AgentCore](<https://devfeed.tech/topics/amazon-bedrock-agentcore.md>), [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [aws-sns](<https://devfeed.tech/topics/aws-sns.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon-api-gateway](<https://devfeed.tech/tags/amazon-api-gateway.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [amazon-bedrock-agentcore](<https://devfeed.tech/tags/amazon-bedrock-agentcore.md>), [amazon-eventbridge](<https://devfeed.tech/tags/amazon-eventbridge.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [amazon-sns](<https://devfeed.tech/tags/amazon-sns.md>), [amazon-vpc](<https://devfeed.tech/tags/amazon-vpc.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [aws-lambda](<https://devfeed.tech/tags/aws-lambda.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [episodic-memory](<https://devfeed.tech/tags/episodic-memory.md>), [learning](<https://devfeed.tech/tags/learning.md>), [luna](<https://devfeed.tech/tags/luna.md>), [memory](<https://devfeed.tech/tags/memory.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [risk](<https://devfeed.tech/tags/risk.md>), [security](<https://devfeed.tech/tags/security.md>), [soc](<https://devfeed.tech/tags/soc.md>), [triage](<https://devfeed.tech/tags/triage.md>)

### AI overview

Greenlight says Luna autonomously resolves more than 90% of up to 4,000 monthly security and risk cases, escalating cases that need human judgment. Built on Amazon Bedrock AgentCore runtime and memory, Luna uses a Tier 1 supervisor to route alerts to more than 30 specialized agents; a Tier 2 supervisor investigates lower-confidence cases across multiple data sources and prepares evidence for review. A learning pipeline records resolutions, deduplicates patterns, and feeds context back to the supervisors; Greenlight attributes the rise from 83% Tier 1 resolution to over 90% to the combined system and learnings. Episodic memory covers 30 days, while long-term memory stores preferences and recurring-case templates. The system also triages dark-web card alerts by checking ownership and activity, deduplicating reports, and cross-referencing customer sessions and support tickets. In regulated workflows, Luna recommends and provides evidence; remediation requires a human or approved policy.

### Source excerpt

Greenlight Financial resolves more than 90% of up to 4,000 monthly security and risk cases without human intervention using Luna, a tiered multi-agent system on Amazon Bedrock AgentCore. This post covers the architecture, episodic memory for continuous learning, and the expansion from SOC operations into financial risk triage.

## The Crisis Is Fatigue: Why Security Needs Autonomous Defense

DevFeed: [The Crisis Is Fatigue: Why Security Needs Autonomous Defense](<https://devfeed.tech/articles/the-crisis-is-fatigue-why-security-needs-autonomous-defense-121024.md>)

Original publisher: [Read original article](<https://hackernoon.com/the-crisis-is-fatigue-why-security-needs-autonomous-defense?source=rss>)

Author: GUARDDOG AI, Inc.

Published: 2026-11-06T07:54:09Z

Content type: opinion

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [network security](<https://devfeed.tech/topics/network-security.md>), [Threat monitoring](<https://devfeed.tech/topics/threat-monitoring.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Endpoint security](<https://devfeed.tech/topics/endpoint-security.md>), [observability](<https://devfeed.tech/topics/observability.md>), [trust](<https://devfeed.tech/topics/trust.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [adversary](<https://devfeed.tech/tags/adversary.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-threat-detection](<https://devfeed.tech/tags/ai-threat-detection.md>), [alarm](<https://devfeed.tech/tags/alarm.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [application](<https://devfeed.tech/tags/application.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-cybersecurity](<https://devfeed.tech/tags/autonomous-cybersecurity.md>), [cybersecurity-alert-fatigue](<https://devfeed.tech/tags/cybersecurity-alert-fatigue.md>), [defense](<https://devfeed.tech/tags/defense.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-cybersecurity](<https://devfeed.tech/tags/enterprise-cybersecurity.md>), [enterprise-technology](<https://devfeed.tech/tags/enterprise-technology.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [guarddog-ai](<https://devfeed.tech/tags/guarddog-ai.md>), [operations](<https://devfeed.tech/tags/operations.md>), [proactive-cybersecurity](<https://devfeed.tech/tags/proactive-cybersecurity.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [technology](<https://devfeed.tech/tags/technology.md>), [threat-containment](<https://devfeed.tech/tags/threat-containment.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The article argues that alert fatigue and manual triage leave security teams too slow to contain lateral movement. It presents GuardDog AI as an agentless defense system: a sidecar alongside workloads in clusters and network sensors for devices that cannot run software agents, providing visibility across multi-cloud, data center and edge environments. It says the system detects behavioral anomalies against a baseline of the organization's own environment and automatically isolates a suspected threat at its network connection point, limiting disruption to the affected device or connection while the rest of the network continues running. The team retains control of decisions after this first response. The text cites a 29-minute average eCrime breakout time from CrowdStrike's 2026 report to explain the urgency. These are the article's product claims; the supplied text does not provide independent evaluation.

### Source excerpt

The crisis in enterprise technology is a crisis of fatigue, with security operations centers overwhelmed by thousands of alerts daily.

## When Everything Else Fails: Why Your Backup Platform Is Your Last Line of Defense

DevFeed: [When Everything Else Fails: Why Your Backup Platform Is Your Last Line of Defense](<https://devfeed.tech/articles/when-everything-else-fails-why-your-backup-platform-is-your-last-line-of-defense-119301.md>)

Original publisher: [Read original article](<https://www.veeam.com/blog/backup-last-line-of-defense.html>)

Author: Salim Ruffin

Published: 2026-11-05T09:47:39Z

Content type: opinion

Language: en

Sources: [Veeam Software Official Blog](<https://devfeed.tech/sources/veeam-software-official-blog.md>)

Topics: [Featured-Backing Up](<https://devfeed.tech/topics/featured-backing-up.md>), [backups](<https://devfeed.tech/topics/backups.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [downtime](<https://devfeed.tech/topics/downtime.md>)

Tags: [account](<https://devfeed.tech/tags/account.md>), [backup](<https://devfeed.tech/tags/backup.md>), [boardroom](<https://devfeed.tech/tags/boardroom.md>), [business](<https://devfeed.tech/tags/business.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [data](<https://devfeed.tech/tags/data.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [defense](<https://devfeed.tech/tags/defense.md>), [first](<https://devfeed.tech/tags/first.md>), [government](<https://devfeed.tech/tags/government.md>), [immutability](<https://devfeed.tech/tags/immutability.md>), [making](<https://devfeed.tech/tags/making.md>), [one](<https://devfeed.tech/tags/one.md>), [operations](<https://devfeed.tech/tags/operations.md>), [platform](<https://devfeed.tech/tags/platform.md>), [ransomware-resilience](<https://devfeed.tech/tags/ransomware-resilience.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [security-operations-center](<https://devfeed.tech/tags/security-operations-center.md>), [shift](<https://devfeed.tech/tags/shift.md>), [software](<https://devfeed.tech/tags/software.md>), [systems](<https://devfeed.tech/tags/systems.md>), [this](<https://devfeed.tech/tags/this.md>), [veeam](<https://devfeed.tech/tags/veeam.md>), [veeam-backup-replication](<https://devfeed.tech/tags/veeam-backup-replication.md>), [veeam-software](<https://devfeed.tech/tags/veeam-software.md>)

### AI overview

The article argues that prevention tools cannot guarantee availability, so organizations need hardened, tested backup and recovery plans before a crisis. It highlights Veeam Backup & Replication capabilities: the Veeam Software Appliance ships with an operating system configured to DISA STIG guidelines; properly configured immutable repositories prevent backup files from being changed or deleted during the defined immutability period, including if the backup server or an administrator account is compromised; and documented offline workflows support disconnected deployments. Veeam can also send security and audit events to external monitoring and response platforms, while its malware detection and monitoring can flag suspicious backup activity and potentially affected restore points. Offline and other deployments remain subject to each organization's security, accreditation, and media-transfer requirements.

### Source excerpt

Whether you are sitting in a boardroom making strategic decisions, managing a security operations center, or working a weekend shift keeping systems alive in a data center, this conversation belongs to every one of us equally. The post When Everything Else Fails: Why Your Backup Platform Is Your Last Line of Defense appeared first on Veeam Software Official Blog.

## Constraint-Aware Evaluation of Reinforcement Learning and Unsupervised Anomaly Detection for Security Operations Alert Triage

DevFeed: [Constraint-Aware Evaluation of Reinforcement Learning and Unsupervised Anomaly Detection for Security Operations Alert Triage](<https://devfeed.tech/articles/constraint-aware-evaluation-of-reinforcement-learning-and-unsupervised-anomaly-detection-for-security-operations-alert-triage-118687.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.05746>)

Author: Radityo Ghifiardi, Rifny Rachman

Published: 2026-11-05T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [actions](<https://devfeed.tech/tags/actions.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alert-triage](<https://devfeed.tech/tags/alert-triage.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [detection](<https://devfeed.tech/tags/detection.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [learning](<https://devfeed.tech/tags/learning.md>), [operations](<https://devfeed.tech/tags/operations.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [triage](<https://devfeed.tech/tags/triage.md>)

### AI overview

The study frames SOC alert triage as choosing alerts to escalate under limited analyst capacity and a benign-escalation constraint. It evaluates an offline reinforcement learning policy trained on alert embeddings, Isolation Forest on operational features, and an autoencoder trained on historical benign alerts. On 74,909 analyst-labeled production alerts with a strict temporal split, the methods are assessed mainly by clusters because the test set contains 503 unique feature patterns. At the main operating point, all meet the constraint; benign-escalation fractions are 0.000 for reinforcement learning, 0.007 for the autoencoder, and 0.041 for Isolation Forest. They prioritize largely different clusters, suggesting complementary coverage. The supplied abstract suggests combined or adaptive deployment may benefit production triage.

### Source excerpt

arXiv:2611.05746v1 Announce Type: new Abstract: Security Operations Centers (SOCs) must identify real threats from large alert streams, but limited analyst capacity means that unnecessary escalations can delay more serious investigations. Existing studies often frame SOC alert triage as classification or anomaly detection, but both views are incomplete for production SOCs, where the central problem is selecting alerts worth escalating under limited capacity, asymmetric error costs, and workload constraints. This paper addresses that gap by formalizing alert triage as a constrained decision problem and evaluating three approaches: an offline reinforcement learning policy trained on alert embeddings, an Isolation Forest model built on operational features, and an autoencoder trained on historical benign alerts. These methods represent different triage assumptions: learning actions from reward, isolating unusual operational patterns, and detecting deviations from benign behavior. We evaluate all methods on 74,909 analyst-labeled production alerts under a strict temporal split and benign-escalation constraint. Because the test set contains 503 unique feature patterns, we primarily rank clusters of similar alerts. At the main operating point, all three methods satisfy the constraint, with the reinforcement learning policy achieving the lowest benign-escalation fraction at 0.000, compared with 0.007 for the autoencoder and 0.041 for Isolation Forest. The methods also prioritize largely different alert clusters, showing complementary rather than redundant coverage. These results suggest that production SOC triage can benefit from constraint-aware evaluation and combined or adaptive deployment of different model families.

## Cycurion Habit verifies threats and validates fixes before closing cases

DevFeed: [Cycurion Habit verifies threats and validates fixes before closing cases](<https://devfeed.tech/articles/cycurion-habit-verifies-threats-and-validates-fixes-before-closing-cases-116588.md>)

Original publisher: [Read original article](<https://www.helpnetsecurity.com/2026/11/04/cycurion-habit/>)

Author: Industry News

Published: 2026-11-04T14:40:23Z

Content type: release

Language: en

Sources: [Help Net Security](<https://devfeed.tech/sources/help-net-security.md>)

Topics: [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>), [Threat Hunting & Intel](<https://devfeed.tech/topics/threat-hunting-intel.md>), [Endpoint Security & XDR](<https://devfeed.tech/topics/endpoint-security-xdr.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-driven-security](<https://devfeed.tech/tags/ai-driven-security.md>), [ai-red-team](<https://devfeed.tech/tags/ai-red-team.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-s3](<https://devfeed.tech/tags/aws-s3.md>), [backup](<https://devfeed.tech/tags/backup.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [disruption](<https://devfeed.tech/tags/disruption.md>), [email-security](<https://devfeed.tech/tags/email-security.md>), [endpoint-protection](<https://devfeed.tech/tags/endpoint-protection.md>), [exploitability](<https://devfeed.tech/tags/exploitability.md>), [identity](<https://devfeed.tech/tags/identity.md>), [industry-news](<https://devfeed.tech/tags/industry-news.md>), [launch](<https://devfeed.tech/tags/launch.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations-center](<https://devfeed.tech/tags/security-operations-center.md>), [threats](<https://devfeed.tech/tags/threats.md>)

### AI overview

Habit establishes behavioral baselines for identities, endpoints and workloads, then combines AI triage with 24/7 human analysts to verify suspected threats, remediate them and validate fixes before closing cases. It connects to existing tools or uses its own agent. An on-demand AI red team, mapped to MITRE ATT&CK, tests the client's environment and retests fixes. Cycurion says unlimited incident response is included without hourly or per-incident fees. The excerpt also lists integrations and support for endpoint protection, email security, and backups for Microsoft 365, Google Workspace and AWS S3. Habit is positioned for existing managed-security clients, public agencies and commercial organizations seeking product capabilities with human oversight.

### Source excerpt

Cycurion has launched Habit, an AI-driven security operations center (SOC) built to identify threats before they disrupt operations. The launch also marks Cycurion's repositioning as a predictive cybersecurity company, built around a newly assembled team of former U.S. Intelligence Community and private-sector cyber experts. Habit doesn't chase alerts. It learns what normal looks like for every identity, device, and workload in a client's environment, then goes after whatever breaks the pattern. It connects to the ... More -> The post Cycurion Habit verifies threats and validates fixes before closing cases appeared first on Help Net Security.

## Brava Raises $15M to Address the Blind Spots in AI-Powered Cybersecurity

DevFeed: [Brava Raises $15M to Address the Blind Spots in AI-Powered Cybersecurity](<https://devfeed.tech/articles/brava-raises-15m-to-address-the-blind-spots-in-ai-powered-cybersecurity-117618.md>)

Original publisher: [Read original article](<https://hackernoon.com/brava-raises-$15m-to-address-the-blind-spots-in-ai-powered-cybersecurity?source=rss>)

Author: Kevin John

Published: 2026-11-04T14:30:03Z

Content type: news

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Detection engineering](<https://devfeed.tech/topics/detection-engineering.md>), [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Security Information and Event Management (SIEM)](<https://devfeed.tech/topics/siem-security.md>), [observability](<https://devfeed.tech/topics/observability.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-attack-simulation](<https://devfeed.tech/tags/ai-attack-simulation.md>), [ai-driven-soc](<https://devfeed.tech/tags/ai-driven-soc.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [attack](<https://devfeed.tech/tags/attack.md>), [brava-15m-seed-round](<https://devfeed.tech/tags/brava-15m-seed-round.md>), [brava-security](<https://devfeed.tech/tags/brava-security.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [detection](<https://devfeed.tech/tags/detection.md>), [integrations](<https://devfeed.tech/tags/integrations.md>), [multi-agent-security-operation](<https://devfeed.tech/tags/multi-agent-security-operation.md>), [platform](<https://devfeed.tech/tags/platform.md>), [raises](<https://devfeed.tech/tags/raises.md>), [security](<https://devfeed.tech/tags/security.md>), [siem](<https://devfeed.tech/tags/siem.md>), [siem-rule-repair](<https://devfeed.tech/tags/siem-rule-repair.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [soc](<https://devfeed.tech/tags/soc.md>), [soc-detection-validation](<https://devfeed.tech/tags/soc-detection-validation.md>), [splunk-and-google-secops](<https://devfeed.tech/tags/splunk-and-google-secops.md>), [telemetry](<https://devfeed.tech/tags/telemetry.md>), [telemetry-gap-analysis](<https://devfeed.tech/tags/telemetry-gap-analysis.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

Brava Security raised $15 million to develop its multi-agent AI platform, which tests attack techniques against an organization's security environment and checks whether the resulting telemetry and SIEM detections expose them. It aims to identify missing logs, broken data feeds, and detection gaps, then help rewrite detections and investigate alongside security teams. The platform is designed to work with existing infrastructure and supports Splunk, Google SecOps, CrowdStrike, and more than 70 other sources and destinations; the company says first insights can arrive 10-15 minutes after onboarding. Brava reports Fortune 100 customers and says funding will support product development and North American expansion. These customer, performance, and funding details are attributed to the company in the supplied article.

### Source excerpt

Brava Security raises $15M to expand its AI-driven SOC platform for attack simulation, detection validation, telemetry gaps, and SIEM integrations.

## CrowdStrike Extends the Agentic SOC to Detect and Stop Agent Insider Threats

DevFeed: [CrowdStrike Extends the Agentic SOC to Detect and Stop Agent Insider Threats](<https://devfeed.tech/articles/crowdstrike-extends-the-agentic-soc-to-detect-and-stop-agent-insider-threats-116046.md>)

Original publisher: [Read original article](<https://www.crowdstrike.com/en-us/blog/crowdstrike-detects-and-stops-agent-insider-threats/>)

Author: Brandon Benke

Published: 2026-11-04T10:11:45.412397Z

Content type: release

Language: en

Sources: [Crowdstrike](<https://devfeed.tech/sources/blog.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-soc](<https://devfeed.tech/tags/agentic-soc.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-security](<https://devfeed.tech/tags/ai-security.md>), [compromise](<https://devfeed.tech/tags/compromise.md>), [crowdstrike](<https://devfeed.tech/tags/crowdstrike.md>), [detection](<https://devfeed.tech/tags/detection.md>), [domain](<https://devfeed.tech/tags/domain.md>), [environment](<https://devfeed.tech/tags/environment.md>), [falcon](<https://devfeed.tech/tags/falcon.md>), [identity](<https://devfeed.tech/tags/identity.md>), [insider](<https://devfeed.tech/tags/insider.md>), [insider-threat](<https://devfeed.tech/tags/insider-threat.md>), [inventory](<https://devfeed.tech/tags/inventory.md>), [investigations](<https://devfeed.tech/tags/investigations.md>), [platform](<https://devfeed.tech/tags/platform.md>), [soc](<https://devfeed.tech/tags/soc.md>), [threats](<https://devfeed.tech/tags/threats.md>)

### AI overview

A trusted AI agent can become an insider threat if a tool is tampered with or its instructions change, while it continues acting under an employee's identity and access. The excerpt says many agentic SOC tools process detections serially with a single LLM and lack memory plus identity, endpoint, and ownership context, limiting their ability to distinguish routine activity from compromise. Standalone AI security tools may inventory agents and flag unusual activity, but the excerpt says they stop at detection and hand response to another tool. CrowdStrike says its Falcon platform is intended to secure agents in the environment and use expert agents in the SOC to investigate and respond at machine speed.

### Source excerpt

CrowdStrike extends the agentic SOC to detect agent insider threats, investigate across every domain, and govern response on the Falcon platform.

## AI in the SOC: Where security teams stand

DevFeed: [AI in the SOC: Where security teams stand](<https://devfeed.tech/articles/ai-in-the-soc-where-security-teams-stand-118947.md>)

Original publisher: [Read original article](<https://www.elastic.co/blog/ai-adoption-in-soc>)

Author: Joe DeFever

Published: 2026-11-04T00:00:00Z

Content type: article

Language: en

Sources: [Elastic Blog](<https://devfeed.tech/sources/elastic-blog-elasticsearch-kibana-and-elk-stack.md>)

Topics: [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [Security](<https://devfeed.tech/topics/security.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>)

Tags: [adoption](<https://devfeed.tech/tags/adoption.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-adoption](<https://devfeed.tech/tags/ai-adoption.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [ai-features](<https://devfeed.tech/tags/ai-features.md>), [ai-for-security](<https://devfeed.tech/tags/ai-for-security.md>), [ai-maturity](<https://devfeed.tech/tags/ai-maturity.md>), [ai-reasoning](<https://devfeed.tech/tags/ai-reasoning.md>), [ai-research](<https://devfeed.tech/tags/ai-research.md>), [ai-soc](<https://devfeed.tech/tags/ai-soc.md>), [ai-tool](<https://devfeed.tech/tags/ai-tool.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [ai-training](<https://devfeed.tech/tags/ai-training.md>), [governance](<https://devfeed.tech/tags/governance.md>), [security](<https://devfeed.tech/tags/security.md>), [soc](<https://devfeed.tech/tags/soc.md>), [teams](<https://devfeed.tech/tags/teams.md>)

### AI overview

A survey of more than 400 security operations professionals found broad AI use and value alongside substantial friction: 96% of teams use AI, 99.8% report benefits, and 85% report negative impacts. General chatbots are used more often than native security-tool AI (75% vs. 66%), with respondents citing broader task coverage, cost and token management, and ease of use. The article links separate "sidecar" tools to context switching, possible sensitive-data exposure, less grounded answers, and added verification work; 50% say most AI outputs need correction or rework. It argues that embedded AI working on security data in place can make reasoning easier to verify, and recommends transparency, editability, data proximity, and human approval for consequential actions. Although 89% favor agentic workflows with oversight and governance, only 18% report a fully developed governance framework. The article presents governance and transparency as prerequisites for expanding automation, with agents doing investigative work and people approving consequential decisions.

### Source excerpt

New data from 400+ security pros reveals the gap between AI SOC adoption and governance, and what an enforced AI governance framework needs.

## Gartner Predicts SOCs Will Have 50% More Engineers Than Analysts by 2028

DevFeed: [Gartner Predicts SOCs Will Have 50% More Engineers Than Analysts by 2028](<https://devfeed.tech/articles/gartner-predicts-socs-will-have-50-more-engineers-than-analysts-by-2028-114229.md>)

Original publisher: [Read original article](<https://www.bleepingcomputer.com/news/security/gartner-predicts-socs-will-have-50-percent-more-engineers-than-analysts-by-2028/>)

Author: Sponsored by Prophet Security

Published: 2026-11-03T15:00:10Z

Content type: article

Language: en

Sources: [BleepingComputer](<https://devfeed.tech/sources/bleepingcomputer.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [alert](<https://devfeed.tech/tags/alert.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [computer-help](<https://devfeed.tech/tags/computer-help.md>), [computer-security](<https://devfeed.tech/tags/computer-security.md>), [computers](<https://devfeed.tech/tags/computers.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [detection](<https://devfeed.tech/tags/detection.md>), [detection-engineering](<https://devfeed.tech/tags/detection-engineering.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [exposure-management](<https://devfeed.tech/tags/exposure-management.md>), [gartner](<https://devfeed.tech/tags/gartner.md>), [handling](<https://devfeed.tech/tags/handling.md>), [infosec](<https://devfeed.tech/tags/infosec.md>), [linux](<https://devfeed.tech/tags/linux.md>), [mac](<https://devfeed.tech/tags/mac.md>), [malware](<https://devfeed.tech/tags/malware.md>), [malware-removal](<https://devfeed.tech/tags/malware-removal.md>), [management](<https://devfeed.tech/tags/management.md>), [more](<https://devfeed.tech/tags/more.md>), [operations](<https://devfeed.tech/tags/operations.md>), [over](<https://devfeed.tech/tags/over.md>), [prophet-security](<https://devfeed.tech/tags/prophet-security.md>), [roles](<https://devfeed.tech/tags/roles.md>), [security](<https://devfeed.tech/tags/security.md>), [security-operations-center](<https://devfeed.tech/tags/security-operations-center.md>), [soc](<https://devfeed.tech/tags/soc.md>), [spyware](<https://devfeed.tech/tags/spyware.md>), [support](<https://devfeed.tech/tags/support.md>), [tech-support](<https://devfeed.tech/tags/tech-support.md>), [technical-support](<https://devfeed.tech/tags/technical-support.md>), [virus](<https://devfeed.tech/tags/virus.md>), [virus-removal](<https://devfeed.tech/tags/virus-removal.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

Gartner forecasts that by 2028 SOCs will have 50% more engineers than analysts, as AI agents reduce routine alert handling. Analysts are expected to shift from triage to validating agent findings, owning cases, choosing responses, and communicating options; junior staff need deliberate practice through worked examples, supervised case ownership, and feedback. Detection engineering expands into SOC engineering, including prompt design, workflow testing, output validation, and software practices such as versioning and rollback. Gartner recommends tracking case accuracy, cycle time, and decision confidence alongside existing metrics, and making continuous exposure management a standing SOC function. The article suggests planning role changes, entry-level training, measurement baselines, and continuous exposure work before rollout. It also describes Prophet Security's product capabilities and reports survey findings, including that 57% of respondents expect team size to stay the same and 9% expect growth. These are forecasts and survey results reported in the supplied excerpt.

### Source excerpt

Gartner predicts that security operations centers will have 50% more engineers than analysts by 2028 as AI agents take over routine alert handling. Prophet Security examines how AI is reshaping SOC roles, from analyst responsibilities and detection engineering to threat hunting and exposure management. [...]

## Your New MTTR Shortcut -- Introducing Related Threat Context

DevFeed: [Your New MTTR Shortcut -- Introducing Related Threat Context](<https://devfeed.tech/articles/your-new-mttr-shortcut-introducing-related-threat-context-113703.md>)

Original publisher: [Read original article](<https://hackernoon.com/your-new-mttr-shortcut-introducing-related-threat-context?source=rss>)

Author: ANY.RUN

Published: 2026-11-03T10:48:31Z

Content type: release

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [Threat Hunting & Intel](<https://devfeed.tech/topics/threat-hunting-intel.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>), [Threat Research](<https://devfeed.tech/topics/threat-research.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [any-run-interactive-sandbox](<https://devfeed.tech/tags/any-run-interactive-sandbox.md>), [any-run-threat-context](<https://devfeed.tech/tags/any-run-threat-context.md>), [any-run-threat-intelligence](<https://devfeed.tech/tags/any-run-threat-intelligence.md>), [attack](<https://devfeed.tech/tags/attack.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [context](<https://devfeed.tech/tags/context.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [industry](<https://devfeed.tech/tags/industry.md>), [intelligence](<https://devfeed.tech/tags/intelligence.md>), [investigations](<https://devfeed.tech/tags/investigations.md>), [iocs](<https://devfeed.tech/tags/iocs.md>), [mssp-threat-analysis](<https://devfeed.tech/tags/mssp-threat-analysis.md>), [mttr](<https://devfeed.tech/tags/mttr.md>), [mttr-reduction](<https://devfeed.tech/tags/mttr-reduction.md>), [net](<https://devfeed.tech/tags/net.md>), [soc](<https://devfeed.tech/tags/soc.md>), [soc-investigation-workflow](<https://devfeed.tech/tags/soc-investigation-workflow.md>), [soc-time-to-triage](<https://devfeed.tech/tags/soc-time-to-triage.md>), [speed](<https://devfeed.tech/tags/speed.md>), [team](<https://devfeed.tech/tags/team.md>), [threat-intelligence](<https://devfeed.tech/tags/threat-intelligence.md>), [threat-intelligence-context](<https://devfeed.tech/tags/threat-intelligence-context.md>), [triage](<https://devfeed.tech/tags/triage.md>)

### AI overview

ANY.RUN's Related Threat Context surfaces an IOC's prior team analyses, industry-associated activity, and ANY.RUN Threat Intelligence in the Interactive Sandbox, helping analysts decide whether to close, investigate, enrich, or escalate a case. Enterprise internal SOCs can access all three context types; MSSPs can access team and TI context, but not industry context. Analysts can follow related IOCs into Threat Intelligence Lookup to investigate linked samples, infrastructure, and behavior. "Never Seen Before" and "Commonly Seen" are prioritization signals, not verdicts, and an industry match is not proof of targeting. The feature can give Tier 2 a more actionable starting point and reduce repeated research, while analysts still need to weigh context alongside behavior and other evidence.

### Source excerpt

Discover how ANY.RUN Related Threat Context connects IOCs to team investigations, industry observations, and threat intelligence to speed up SOC triage.

## StateFence: Task-State Protection Against Evidence-Chain Contamination in Blue-Team LLM Agents

DevFeed: [StateFence: Task-State Protection Against Evidence-Chain Contamination in Blue-Team LLM Agents](<https://devfeed.tech/articles/statefence-task-state-protection-against-evidence-chain-contamination-in-blue-team-llm-agents-114081.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.02605>)

Author: Haowen Dai, Alejandro Guerra-Manzanares

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [LLM security](<https://devfeed.tech/topics/llm-security.md>), [prompt injection](<https://devfeed.tech/topics/prompt-injection.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>)

Tags: [alert-triage](<https://devfeed.tech/tags/alert-triage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [defense](<https://devfeed.tech/tags/defense.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [indirect-prompt-injection](<https://devfeed.tech/tags/indirect-prompt-injection.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [security-operations](<https://devfeed.tech/tags/security-operations.md>), [soc](<https://devfeed.tech/tags/soc.md>)

### AI overview

StateFence addresses indirect prompt injection in blue-team LLM agents by protecting task-critical state as untrusted evidence moves through SOC workflows. It treats severity labels, incident narratives, response policies, and IOC sets as protected states, maps attacks as explicit, stealth, multi-hop, or persistent, and records observable propagation traces. Defenses are routed according to the state under attack because suspicious text can be malicious control in one task yet essential evidence in another--for example, during IOC extraction. In controlled SOC and public-SOC validation cases, the framework reportedly lowered attack success and downstream failures while preserving task utility and avoiding indicator loss from task-mismatched filtering.

### Source excerpt

arXiv:2611.02605v1 Announce Type: new Abstract: Indirect prompt injection creates a distinctive risk for blue-team Large Language Model (LLM) agents in Security Operations Centers (SOCs). Malicious instructions can be embedded in telemetry, tickets, retrieval results, or memory and then propagate through alert triage, incident summarization, response recommendation, and Indicator of Compromise (IOC) extraction as locally plausible intermediate decisions. We characterize this failure mode as evidence-chain contamination of protected task state. The main challenge is task dependent: a suspicious string may be malicious control text for summarization but essential evidence for IOC extraction. To address this challenge, we propose StateFence, a task-aware evaluation and defense framework that treats severity labels, incident narratives, response policies, and IOC sets as protected states. StateFence maps indirect prompt-injection attacks into explicit, stealth, multi-hop, and persistent SOC evidence attacks, records observable propagation traces, and routes defenses according to the attacked state. Across controlled SOC cases and public-SOC validation cases, StateFence lowers attack success and reduces failures that appear only after evidence is poisoned, while preserving task utility and avoiding the indicator loss caused by task-mismatched filtering. Our findings indicate that robust blue-team LLM agents require mechanisms that preserve task-critical state under untrusted evidence.

## PHP2Shell: How Cisco Helps Customers Respond to AI-Accelerated Threats

DevFeed: [PHP2Shell: How Cisco Helps Customers Respond to AI-Accelerated Threats](<https://devfeed.tech/articles/php2shell-how-cisco-helps-customers-respond-to-ai-accelerated-threats-115483.md>)

Original publisher: [Read original article](<https://blogs.cisco.com/security/php2shell-how-cisco-helps-customers-respond-to-ai-accelerated-threats/>)

Author: Rick Miles

Published: 2026-11-02T18:24:13Z

Content type: article

Language: en

Sources: [Security @ Cisco](<https://devfeed.tech/sources/security-cisco-blogs.md>)

Topics: [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>), [Cisco](<https://devfeed.tech/topics/cisco.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Cisco Talos](<https://devfeed.tech/topics/cisco-talos.md>), [Resilience](<https://devfeed.tech/topics/resilience.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [addition](<https://devfeed.tech/tags/addition.md>), [advanced](<https://devfeed.tech/tags/advanced.md>), [ai](<https://devfeed.tech/tags/ai.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [all](<https://devfeed.tech/tags/all.md>), [apply](<https://devfeed.tech/tags/apply.md>), [assets](<https://devfeed.tech/tags/assets.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [block](<https://devfeed.tech/tags/block.md>), [can](<https://devfeed.tech/tags/can.md>), [cisco](<https://devfeed.tech/tags/cisco.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [customers](<https://devfeed.tech/tags/customers.md>), [firewall](<https://devfeed.tech/tags/firewall.md>), [next](<https://devfeed.tech/tags/next.md>), [php2shell](<https://devfeed.tech/tags/php2shell.md>), [resources](<https://devfeed.tech/tags/resources.md>), [security](<https://devfeed.tech/tags/security.md>), [splunk](<https://devfeed.tech/tags/splunk.md>), [threats](<https://devfeed.tech/tags/threats.md>)

### AI overview

Cisco says it had PHP2Shell exploit coverage before disclosure, day-zero signatures, and Splunk detections for post-exploit activity. In customer firewall telemetry, SnortML identified and blocked a novel remote-code-execution attempt less than 24 hours after disclosure by matching patterns learned from other attacks. Cisco Firewall software 10.1 includes SnortML; it is enabled by default for customers using the security over connectivity policy. Snort signatures and machine learning block suspicious traffic, while Splunk links firewall alerts to more than 10 detections for PHP2Shell exploitation and post-exploitation activity, including Cisco Firewall-specific detections. The text also points customers to Talos guidance, IPS signatures, SnortML learning resources, Splunk content, and a firewall test drive. It presents patching as the first step and SnortML as a temporary defense while teams patch.

### Source excerpt

The following resources address the coverage available for PHP2Shell today, as well as ways Cisco customers can begin to increase resilience for whatever AI-driven threat comes next.

## What Makes a Good Security Agent? 5 Lessons from the AgentWorks Hackathon

DevFeed: [What Makes a Good Security Agent? 5 Lessons from the AgentWorks Hackathon](<https://devfeed.tech/articles/what-makes-a-good-security-agent-5-lessons-from-the-agentworks-hackathon-112222.md>)

Original publisher: [Read original article](<https://www.crowdstrike.com/en-us/blog/five-lessons-from-agentworks-hackathon/>)

Author: Lucia Stanham

Published: 2026-11-02T13:52:38.384581Z

Content type: article

Language: en

Sources: [Crowdstrike](<https://devfeed.tech/sources/blog.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [Threat monitoring](<https://devfeed.tech/topics/threat-monitoring.md>), [Security Information and Event Management (SIEM)](<https://devfeed.tech/topics/siem-security.md>), [Hackathon](<https://devfeed.tech/topics/hackathon.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-security](<https://devfeed.tech/tags/agent-security.md>), [agentic-soc](<https://devfeed.tech/tags/agentic-soc.md>), [ai-reasoning](<https://devfeed.tech/tags/ai-reasoning.md>), [hackathon](<https://devfeed.tech/tags/hackathon.md>), [securing-ai](<https://devfeed.tech/tags/securing-ai.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

The strongest security agents target recurring work teams struggle to complete, use AI judgment where needed while delegating fixed tasks to deterministic tools, and include measurable guardrails and human review. Examples include an agent that inventories unapproved MCP servers using SIEM queries and redaction, an MFA reset investigator that scores evidence deterministically and requires approval for containment, and an agent that tests its own proposed detection exclusions for bypasses. The article also describes query validation, structured threat hunting, and investigation agents. AgentWorks traces help builders diagnose ignored errors, empty results mistaken for "none," unnecessary chaining, and unused knowledge bases. The account draws lessons from hackathon submissions and examples; it does not establish that every agent performs equally well in production.

### Source excerpt

The first-ever AgentWorks Hackathon invited security teams to submit agents they built on Charlotte AI AgentWorks to solve a persistent bottleneck and function as a trusted extension of the team. Read about a few that stood out.

## From Reactive to Proactive: Empowering SOC Teams with the Wiz Red and Blue Agents

DevFeed: [From Reactive to Proactive: Empowering SOC Teams with the Wiz Red and Blue Agents](<https://devfeed.tech/articles/from-reactive-to-proactive-empowering-soc-teams-with-the-wiz-red-and-blue-agents-112527.md>)

Original publisher: [Read original article](<https://www.wiz.io/blog/soc-alerts-red-and-blue-agents>)

Author: Shaked Rotlevi

Published: 2026-11-02T13:13:46Z

Content type: article

Language: en

Sources: [Wiz](<https://devfeed.tech/sources/wiz-blog-rss-feed.md>)

Topics: [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>), [Security Operations Center](<https://devfeed.tech/topics/security-operations-center.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-threat](<https://devfeed.tech/tags/ai-threat.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [breach](<https://devfeed.tech/tags/breach.md>), [era](<https://devfeed.tech/tags/era.md>), [gives](<https://devfeed.tech/tags/gives.md>), [machine](<https://devfeed.tech/tags/machine.md>), [proactive](<https://devfeed.tech/tags/proactive.md>), [product](<https://devfeed.tech/tags/product.md>), [reactive](<https://devfeed.tech/tags/reactive.md>), [red](<https://devfeed.tech/tags/red.md>), [soc](<https://devfeed.tech/tags/soc.md>), [speed](<https://devfeed.tech/tags/speed.md>), [teams](<https://devfeed.tech/tags/teams.md>), [threats](<https://devfeed.tech/tags/threats.md>), [with](<https://devfeed.tech/tags/with.md>), [wiz](<https://devfeed.tech/tags/wiz.md>), [wiz-cloud](<https://devfeed.tech/tags/wiz-cloud.md>)

### AI overview

Wiz's Red and Blue Agents are presented as a way for SOC teams to act before and during attacks: Red Agent continuously tests the attack surface and turns validated exploitable risks--including logic flaws signature-based scanners may miss--into pre-breach alerts. Findings include exploit proof, reproduction steps, Security Graph blast-radius context, and remediation guidance routed by Green Agent. The suggested response is to scope possible prior use, contain with stopgap measures, hand off the fix, then have Red Agent retest. Blue Agent investigates active threats using Security Graph, code-to-cloud, runtime, and detection-time forensic evidence from Wiz Sensor; it shows its questions and evidence so analysts can assess its verdict. Wiz Workflows can trigger escalation, notifications, or containment based on verdict and confidence. The source describes these capabilities but provides no independent performance evaluation.

### Source excerpt

Red Agent gives SOC teams validated pre-breach alerts, while Blue Agent investigates active threats at machine speed, helping teams stay ahead in the AI threat era.

[Next page](<https://devfeed.tech/topics/security-operations-center.md?cursor=WyIyMDI2LTExLTAyVDEzOjEzOjQ2KzAwOjAwIiwgImFjYTgxOWUzLTI2ZjUtNDgzMS1hOGU2LWQ1ZDVmOTY5M2VkNyJd>)