<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Harikrishnan P on Medium]]></title>
        <description><![CDATA[Stories by Harikrishnan P on Medium]]></description>
        <link>https://medium.com/@harikrishnan.p097?source=rss-a542fa5a502b------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*_NfkylHm_9Z7yAF0</url>
            <title>Stories by Harikrishnan P on Medium</title>
            <link>https://medium.com/@harikrishnan.p097?source=rss-a542fa5a502b------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 06:51:09 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@harikrishnan.p097/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Operating an Active Directory SOC Lab: Detection Validation, Investigation & Incident Response]]></title>
            <link>https://medium.com/@harikrishnan.p097/operating-an-active-directory-soc-lab-detection-validation-investigation-incident-response-dd48c06fc5d8?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/dd48c06fc5d8</guid>
            <category><![CDATA[active-directory]]></category>
            <category><![CDATA[wazuh]]></category>
            <category><![CDATA[incident-response]]></category>
            <category><![CDATA[detection-engineering]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Sun, 04 Oct 2026 08:51:16 GMT</pubDate>
            <atom:updated>2026-10-04T08:51:16.636Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*drcR1SBXLLYgLxjygPYEFg.png" /></figure><h4>How I validated 42 custom detections, connected alerts to investigation and response, and documented what I learned from building the lab.</h4><p>This is Part 2 of the SOC Active Directory Detection and Response Lab series. New here? Start with <a href="https://medium.com/@harikrishnan.p097/building-an-active-directory-soc-lab-detection-engineering-threat-hunting-incident-response-6f0f6c10cafd">Part 1: Building an Active Directory SOC Lab: Detection Engineering, Threat Hunting &amp; Incident Response with Wazuh</a></p><p>Part 1 covered the build: Windows and Sysmon telemetry, 42 custom Wazuh detections, MITRE ATT&amp;CK mapping, and SOC dashboards. But an alert is only a starting point. An analyst still has to work out what happened, how far it reached, and what to do about it. This part covers how I validated the detections, how an alert becomes an investigation, the seven incident response playbooks, and what I learned.</p><h4>In this article</h4><ul><li>How I validated the 42 detections</li><li>From detection to incident response</li><li>Worked example: investigating a DET-011 alert</li><li>Connecting the SOC workflow</li><li>Lessons learned</li><li>Inside the repository</li><li>What this lab demonstrates</li><li>Conclusion</li></ul><h3>How I Validated the 42 Detections</h3><p>Writing a detection rule is only the beginning.</p><p>A rule can load successfully, match the expected event fields, and still fail to provide useful detection coverage in practice. To make the detections meaningful, I needed to test them against the behaviors they were designed to identify.</p><p>For this project, all 42 custom Wazuh detections were tested and validated in the isolated lab environment.</p><h4><strong>The Validation Process</strong></h4><p>The validation process followed a repeatable workflow:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wlvWDFP5P-th5n11euVCRA.png" /><figcaption>Validation process</figcaption></figure><p>The objective was to verify the complete path from</p><blockquote><strong>security behavior → telemetry → detection → alert</strong></blockquote><p><strong>1. Generate Controlled Activity</strong></p><p>For each detection, I generated controlled activity designed to reproduce the behavior the rule was intended to detect.</p><p>Depending on the scenario, activity was generated from Kali Linux, CLIENT01, or DC01 within the isolated lab.</p><p>This provided a controlled way to test different categories of activity, including authentication failures, account changes, privilege-related activity, credential attacks, discovery, lateral movement, persistence, defense evasion, and impact.</p><p><strong>2. Verify the Expected Telemetry</strong></p><p>After generating the activity, I first verified that the expected Windows Security Event or Sysmon telemetry was actually produced.</p><p>This step was important because a detection cannot reliably identify behavior if the underlying telemetry is missing, incomplete, or different from what the rule expects.</p><p>The validation therefore included checking:</p><ul><li>Whether the expected event was generated</li><li>Whether the relevant fields were present</li><li>Whether the Wazuh agent collected the event</li><li>Whether the event reached the Wazuh manager</li><li>Whether the telemetry contained enough context for the detection</li></ul><p><strong>3. Verify the Wazuh Alert</strong></p><p>Once the telemetry reached Wazuh, I verified that the corresponding custom detection generated an alert.</p><p>This allowed me to confirm that the complete detection path was working:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bEYgt_TW2XD8lVfWZg-vCQ.png" /><figcaption>Detection path</figcaption></figure><p>The alert was then checked against the expected detection behavior rather than simply being treated as proof that the rule worked.</p><p><strong>4. Tune the Detection</strong></p><p>Testing also exposed cases where detection logic needed refinement.</p><p>Detection tuning involved reviewing the telemetry and rule conditions to determine whether the detection was:</p><ul><li>Matching the intended behavior</li><li>Providing sufficient context</li><li>Producing unnecessary noise</li><li>Too broad or too restrictive</li><li>Triggering consistently against the expected activity</li></ul><p>This iterative process helped improve the quality and usefulness of the detections.</p><p>The goal was not simply to maximize the number of alerts, but to produce alerts that provided meaningful security context to an analyst.</p><p><strong>5. Clean Up After Testing</strong></p><p>After each controlled test, the environment was cleaned up where necessary.</p><p>This helped prevent test activity from affecting subsequent validation exercises and kept the lab in a predictable state.</p><p>Maintaining a controlled environment was particularly important when testing account changes, privilege modifications, persistence mechanisms, or other activities that could affect later tests.</p><p><strong>6. Capture and Document the Results</strong></p><p>Validation evidence was documented throughout the testing process.</p><p>This included recording the tested detection, the expected behavior, the telemetry generated, the resulting Wazuh alert, and the validation outcome.</p><p>The project also documents the relationship between detections and:</p><ul><li>Windows / Sysmon event IDs</li><li>MITRE ATT&amp;CK techniques</li><li>Incident response playbooks</li><li>Detection coverage</li></ul><p>This created a traceable connection between what was tested, what telemetry supported it, what detection identified it, and how an analyst could respond to it.</p><p><strong>Validation Was More Than “The Rule Fired”</strong></p><p>One of the most important lessons from the validation process was that a detection should not be considered complete simply because the rule loads successfully or generates an alert.</p><p>The real question is:</p><blockquote><strong>Does the detection reliably identify the behavior it was designed to detect and provide enough context for an analyst to investigate it?</strong></blockquote><p>That requires testing the complete chain:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DlmrioQAaHDmWjOEWYr6Lw.png" /><figcaption>Detection-to-response chain</figcaption></figure><p>By validating the detections against controlled activity, reviewing the underlying telemetry, tuning the rules, and documenting the results, I was able to treat detection engineering as an iterative process rather than a one-time configuration task.</p><p>This validation process also helped identify areas where detection coverage could be improved — an important part of building a SOC that can evolve as new behaviors and investigation requirements emerge.</p><h3>From Detection to Incident Response</h3><p>A detection tells an analyst that something suspicious may have happened. But a detection by itself does not answer the more important questions:</p><ul><li>What happened?</li><li>How serious is it?</li><li>Which systems and accounts are affected?</li><li>What should be investigated next?</li><li>How should the activity be contained?</li><li>What needs to be recovered or monitored afterward?</li></ul><p>This is where incident response playbooks become useful.</p><p>In this project, I created seven incident response playbooks designed to connect detection activity with a structured investigation and response process.</p><h4><strong>The Incident Response Lifecycle</strong></h4><p>The response lifecycle follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e1Z8Mw71vmf86HnA7K4OEg.png" /><figcaption>Incident response lifecycle</figcaption></figure><p>Each playbook moves through the same stages: triage, investigation and correlation, incident assessment, containment, eradication, recovery and validation, and closure.</p><p>The goal was not to create generic incident-response documentation, but to connect the playbooks to the types of security activity that the lab could actually detect.</p><h4><strong>Connecting Detection to Investigation</strong></h4><p>A SOC incident rarely consists of a single alert.</p><p>For example, an account compromise might produce several related signals:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qT4TIPpvTUcKMR47J8HuEQ.png" /><figcaption>Detections-to-investigation flow</figcaption></figure><p>Individually, these events may require investigation. When correlated by user, host, source, time, and activity, they can provide a much clearer picture of what happened.</p><p>This is an important part of the project: the detections are not treated as isolated alerts. They provide evidence that an analyst can use to build an incident timeline and determine the appropriate response.</p><h4><strong>The Seven Incident Response Playbooks</strong></h4><p>I created seven playbooks covering the major attack scenarios represented in the lab:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/264f72339684612492df6a59e1377ce5/href">https://medium.com/media/264f72339684612492df6a59e1377ce5/href</a></iframe><ul><li><strong>IR-001: Credential Attack &amp; Account Compromise</strong> (DET-001 to DET-006)</li><li><strong>IR-002: Active Directory Account &amp; Privilege Compromise</strong> (DET-007 to DET-016)</li><li><strong>IR-003: Kerberos &amp; Credential Theft Attack</strong> (DET-017 to DET-024)</li><li><strong>IR-004: Active Directory Discovery &amp; Reconnaissance</strong> (DET-025 to DET-027)</li><li><strong>IR-005: Lateral Movement &amp; Remote Execution</strong> (DET-028 to DET-032)</li><li><strong>IR-006: Persistence &amp; Defense Evasion</strong> (DET-033 to DET-041)</li><li><strong>IR-007: AD Destructive &amp; Impact Activity</strong> (DET-042)</li></ul><p>I did not reproduce the complete procedures for all seven playbooks here. The full playbooks, including their investigation and response procedures, are documented in the project repository.</p><h4><strong>From Alert to Incident</strong></h4><p>The relationship between detection and response can be represented as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZnhVtMpF3Iw3DXPHGNO-MA.png" /><figcaption>Alert-to-incident workflow</figcaption></figure><p>For example, a suspicious privileged-group modification could initially trigger a detection based on an Active Directory security event.</p><p>The analyst can then investigate:</p><ul><li>Which account was modified?</li><li>Who performed the change?</li><li>Which group was affected?</li><li>When did the change occur?</li><li>Was the account already showing suspicious authentication activity?</li><li>Were there related endpoint events?</li><li>Are there other affected systems or accounts?</li></ul><p>If the investigation indicates that the activity is part of a larger compromise, the relevant response playbook provides the next steps for containment and remediation.</p><p>This creates a complete operational chain:</p><blockquote>Detect → Investigate → Respond</blockquote><p>rather than stopping at:</p><blockquote>Alert generated → Investigation ends</blockquote><h3>Worked Example: Investigating a DET-011 Alert</h3><p>The questions above are easier to follow with a concrete case, so here is one from the lab, using the Domain Admins detection (DET-011).</p><p><strong>The Scenario</strong></p><p>To validate the detection, I signed in as <strong>CORP\Administrator</strong> and added a test account, <strong>soctest01</strong>, to the Domain Admins group.</p><p>This kind of change is routine during real provisioning and critical when it isn’t authorized, which is why DET-011 always deserves a look.</p><p><strong>The Alert</strong></p><p>Windows logged Event ID 4728 on the domain controller. Wazuh base rule 60103 matched it, and custom rule 100110 raised a level 12 alert:</p><blockquote>“DET-011 Member Added to Domain Admins: soctest01”</blockquote><p><strong>Quick facts</strong></p><ul><li><strong>Made the change:</strong> CORP\Administrator</li><li><strong>Account added:</strong> soctest01</li><li><strong>Group:</strong> Domain Admins</li><li><strong>Windows event:</strong> 4728</li><li><strong>Wazuh rules:</strong> 60103 → 100110</li><li><strong>Detection:</strong> DET-011 (level 12)</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*daGxS8Hj_bxlXRa5ngj3zQ.png" /><figcaption>DET-011 validation chain</figcaption></figure><p>The alert already answers the first questions an analyst has: the group is Domain Admins, and the account that received the privileges is soctest01. Here is how the investigation continues.</p><p><strong>Investigating the Alert</strong></p><p><strong>1. Who made the change?</strong></p><p>The 4728 event records the account that performed the change in its subject fields. Here, the subject is CORP\Administrator.</p><p>In a real environment, I would also check where the change came from, and whether the built-in Administrator account should be making changes at all. Most organizations expect named admin accounts or privileged-access tooling instead.</p><p><strong>2. Was it expected?</strong></p><p>I look for a matching change ticket or privileged access request. In the lab this was a planned test, so it was expected.</p><p>In a real environment, an unapproved change would warrant investigation and escalation according to the organization’s incident-response process, even when the account is a known admin.</p><p><strong>3. What do we know about the added account?</strong></p><p>An account that was just created, recently enabled, or recently given a new password is the highest-risk case. For soctest01, I would review its earlier events:</p><ul><li>Account creation (Event 4720, DET-007)</li><li>Password reset (DET-008)</li><li>Failed or unusual logons</li></ul><p><strong>4. What did the added account do afterward?</strong></p><p>Once an account holds Domain Admins rights, I check for:</p><ul><li>Interactive logons to domain controllers</li><li>Replication activity that looks like DCSync</li><li>GPO changes</li><li>New scheduled tasks</li></ul><p>Sysmon process-creation events show what ran on the host.</p><p><strong>5. What else did the subject account do around the same time?</strong></p><p>I review other group changes, account creation, and logon locations for CORP\Administrator in the minutes before and after.</p><p><strong>6. Decide and respond.</strong></p><ul><li><strong>If the change is authorized:</strong> document the approval and close the alert.</li><li><strong>If it is suspicious:</strong> remove soctest01 from Domain Admins, escalate to the incident response lead, and follow the Active Directory Account &amp; Privilege Compromise playbook (IR-002).</li><li>If compromise is suspected: restrict or disable the affected or compromised account(s), isolate affected systems where appropriate, and review Active Directory for additional persistence, such as extra accounts or GPO changes.</li></ul><p><strong>Cleanup</strong></p><p>After the test, I removed soctest01 from Domain Admins. Windows logged this as Event ID 4729, again with CORP\Administrator as the subject.</p><p>DET-011 matches additions (4728) only, so the removal doesn’t raise a DET-011 alert. It is still a useful event to look for when building a timeline.</p><p>One alert tells you something happened. The pivots (who, when, from where, doing what) turn it into a timeline, and the playbook tells you what to do with it.</p><p><strong>Why Incident Response Playbooks Matter</strong></p><p>Detection rules answer:</p><blockquote>“What suspicious activity occurred?”</blockquote><p>Investigation answers:</p><blockquote>“What actually happened?”</blockquote><p>Incident response answers:</p><blockquote>“What should we do about it?”</blockquote><p>Keeping these stages connected makes the SOC workflow more consistent and repeatable.</p><p>It also allows multiple detections to contribute evidence to a single incident instead of treating every alert as an isolated event.</p><p>For example, a potential account compromise could involve authentication failures, a successful login, privilege changes, and suspicious endpoint activity. Each detection provides a piece of evidence, while the investigation brings those pieces together into a larger incident timeline.</p><h4>Full Incident Response Documentation</h4><p>The complete seven incident response playbooks are in the repository’s <a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab/tree/main/incident-response">incident-response folder</a>.</p><p>They provide the detailed operational procedures behind the response lifecycle described in this section, while the Medium article focuses on the overall SOC architecture and workflow.</p><p><strong>GitHub Repository:</strong><br><a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab">https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab</a></p><p>With detection engineering, dashboards, threat hunting, attack simulation, and incident response now connected, the project moves from individual security capabilities toward a complete SOC workflow:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bN9XucTkJqNpk1L72nflmQ.png" /><figcaption>SOC workflow</figcaption></figure><p>This connection between detection and response is what turns security monitoring into an operational SOC process.</p><h3>Connecting the SOC Workflow</h3><p>Part 1 covered the individual components. Here is how they fit together into one workflow, from raw events to response.</p><p><strong>The End-to-End SOC Workflow</strong></p><p>The workflow implemented in this project can be represented as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9sdMEKkVG_nDlxBfTZdMqg.png" /><figcaption>End-to-end SOC workflow</figcaption></figure><p><strong>1. Telemetry Collection</strong></p><p>The process begins with security telemetry generated by the Windows environment.</p><p>Windows Security Events, Active Directory activity, and Sysmon telemetry provide visibility into authentication, account changes, process activity, network connections, privilege-related activity, and other endpoint behavior.</p><p>This telemetry gives the SOC the evidence required to identify potentially suspicious activity.</p><p><strong>2. Centralized Monitoring with Wazuh</strong></p><p>The telemetry is collected and centralized through Wazuh.</p><p>Wazuh provides the platform for log collection, analysis, custom detection rules, alerting, and security monitoring across the lab environment.</p><p>Instead of investigating individual systems separately, the SOC can work with centralized security data.</p><p><strong>3. Detection Engineering</strong></p><p>The next layer is the 42 custom Wazuh detections developed for the project.</p><p>These detections identify specific security behaviors across areas such as:</p><ul><li>Authentication</li><li>Account Management</li><li>Privilege Escalation</li><li>Credential Access &amp; Kerberos</li><li>Discovery &amp; Reconnaissance</li><li>Lateral Movement</li><li>Persistence &amp; Defense Evasion</li><li>Impact</li></ul><p>The detections transform raw telemetry into security-relevant alerts that an analyst can investigate.</p><p><strong>4. SOC Dashboards</strong></p><p>The alerts and security activity are then presented through seven SOC dashboards.</p><p>The dashboards provide different perspectives of the environment, including authentication activity, Active Directory changes, endpoint activity, threat hunting, MITRE ATT&amp;CK coverage, and detection validation.</p><p>This makes it easier for an analyst to identify patterns, investigate anomalies, and understand what is happening across the environment.</p><p><strong>5. Threat Hunting</strong></p><p>Not every investigation has to begin with a high-severity alert.</p><p>Threat hunting provides a way to proactively search the available telemetry for suspicious patterns, relationships, or activity that may not have immediately generated a dedicated alert.</p><p>In this project, threat hunting is connected to the same telemetry and detection ecosystem used for alert-driven investigations.</p><p><strong>6. Investigation</strong></p><p>Once suspicious activity is identified, the analyst moves from detection into investigation.</p><p>The investigation process involves building context around the activity:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5hTqNb0yTjluUE_TlJBN2w.png" /><figcaption>Investigation process</figcaption></figure><p>This is where individual detections become more valuable when viewed together.</p><p>Multiple alerts can provide different pieces of evidence that help establish an incident timeline and determine the scope of the activity.</p><p><strong>7. Incident Response</strong></p><p>When the investigation identifies an incident requiring action, the appropriate incident response playbook can be used.</p><p>The project contains seven response playbooks, from credential attacks and Active Directory privilege compromise through Kerberos attacks, discovery, lateral movement, persistence and defense evasion, and destructive activity.</p><p>These playbooks connect investigation findings to structured response actions such as containment, eradication, recovery, and lessons learned.</p><h4>The Complete SOC Lifecycle</h4><p>Putting everything together gives the project its overall operational model:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ENsuJ5YxtsDQtD8qI1jhbg.png" /><figcaption>SOC lifecycle</figcaption></figure><p>The final step is particularly important.</p><p>Incident response and investigation can reveal gaps in existing detections. Those findings can then be used to tune detection logic, improve telemetry collection, refine dashboards, and strengthen future investigations.</p><p>This creates a continuous improvement cycle rather than a one-time detection process.</p><h4>Putting the Project Together</h4><p>The architecture can therefore be viewed as a connected SOC pipeline:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*f3Wl4tMVgMKQe6QBEKntVA.png" /><figcaption>SOC architecture</figcaption></figure><p>This is the main idea behind the project.</p><p>The individual components are useful on their own, but their real value comes from connecting them into an end-to-end security operations workflow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bN9XucTkJqNpk1L72nflmQ.png" /><figcaption>SOC workflow</figcaption></figure><p>The project was designed around a simple principle:</p><blockquote><em>A SOC should not stop at detecting suspicious activity. It should provide a path from detection to investigation and, when necessary, to response.</em></blockquote><p>That end-to-end perspective was one of the main goals of building this lab.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b6lbXFjaDrgnxUdVHNRvQA.png" /><figcaption>Detection, investigation and response workflow</figcaption></figure><p>For me, this was one of the most important outcomes of the project. Instead of learning detection, dashboards, threat hunting, and incident response as separate topics, I was able to connect them into one practical security operations workflow.</p><h3>Lessons Learned</h3><p>Building this lab taught me that a functional SOC environment is about much more than collecting logs and writing detection rules.</p><p>The most valuable lessons came from the parts that did not work perfectly the first time — missing telemetry, noisy alerts, detection tuning, and the need to connect individual alerts to a larger investigation.</p><p><strong>1. Detection Quality Depends on Telemetry</strong></p><p>One of the clearest lessons was that a detection is only as reliable as the telemetry behind it.</p><p>While validating the detections, I had to verify not only whether a rule matched, but whether the expected Windows or Sysmon event was actually being generated, collected, and forwarded to Wazuh.</p><p>This made telemetry a fundamental part of detection engineering rather than something that happens in the background.</p><p><strong>In this lab:</strong> DET-011 only works if Windows actually records Event 4728 and Wazuh’s base rule 60103 picks it up. If either link is missing, the custom rule never fires and nothing in the rule itself looks broken. Much of my validation time went into exactly this kind of troubleshooting: checking telemetry fields, rule matching, and alert generation before touching the rule logic.</p><p><strong>2. An Alert Needs Investigation Context</strong></p><p>A detection can tell you that something suspicious happened, but that does not necessarily tell you what happened.</p><p>During the project, I found that useful investigation requires additional context — the affected user, system, source, timing, related events, and activity before and after the alert.</p><p>This changed how I thought about detection design. The goal is not simply to trigger an alert, but to provide an analyst with enough information to investigate it.</p><p><strong>In this lab:</strong> The DET-011 alert names the account that was added to Domain Admins, but the account that made the change lives in the event’s subject fields. An analyst has to look past the alert description to find it, which is why the worked example above starts by asking “who made the change?”</p><p><strong>3. More Alerts Do Not Always Mean Better Detection</strong></p><p>Testing also showed why alert volume needs to be considered carefully.</p><p>Some security events can occur legitimately in an Active Directory environment. A detection that is too broad can therefore generate unnecessary alerts and make it harder for an analyst to identify genuinely suspicious activity.</p><p>This reinforced the importance of balancing detection coverage with signal quality.</p><p><strong>In this lab:</strong> DET-017 (Kerberoasting) is deliberately broad: it fires on Kerberos service-ticket requests (Event 4769). That gives maximum visibility in a small lab, but in a production domain it would flood an analyst with alerts, since legitimate services request tickets constantly. A real version would add filters for RC4 encryption, machine accounts, and request volume.</p><p><strong>4. Detection Tuning Is an Iterative Process</strong></p><p>I initially approached detection development as a process of writing a rule and testing whether it triggered.</p><p>In practice, validation made the process more iterative:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*af45f3k6Ta1r_2WvQjeIng.png" /><figcaption>Detection tuning</figcaption></figure><p>Reviewing the actual telemetry and alert output helped identify where detection logic needed to be refined.</p><p>This became one of the most practical lessons from the project: detection engineering does not really end when the rule is written.</p><p><strong>In this lab:</strong> One result of tuning was layering. DET-001 simply tags every failed Windows logon (Event 4625), and DET-002 and DET-006 build the brute-force and same-source thresholds on top of it. Keeping the base rule simple made the threshold rules easier to adjust.</p><p><strong>5. MITRE ATT&amp;CK Helps Organize Detection Coverage</strong></p><p>Mapping the detections to MITRE ATT&amp;CK gave the project a clearer structure.</p><p>Instead of looking at 42 detections as an isolated collection of rules, I could view them in terms of attacker behaviors and techniques.</p><p>This also made it easier to identify which areas of the attack lifecycle were covered and where additional detection opportunities could exist.</p><p><strong>In this lab:</strong> Mapping also works as a quality check. While preparing this write-up, I found that one rule was tagged with the wrong technique (File Deletion, T1070.004, instead of Clear Windows Event Logs, T1070.001) and corrected it in both the rule and the documentation. A mapping is only useful if it is accurate.</p><p><strong>6. Dashboards Improve Operational Visibility</strong></p><p>The seven dashboards changed how I interacted with the telemetry.</p><p>Rather than manually examining individual events, dashboards provided different perspectives of authentication activity, Active Directory changes, endpoint behavior, detection activity, threat hunting, and ATT&amp;CK coverage.</p><p>The lesson was not that dashboards replace investigation, but that good visualization can help an analyst identify where investigation should begin.</p><p><strong>7. Response Playbooks Turn Detection into Action</strong></p><p>Another important realization was that detection is only one part of the SOC workflow.</p><p>The seven incident response playbooks helped connect an alert to a structured investigation and response process.</p><p>This created a clearer relationship between:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bN9XucTkJqNpk1L72nflmQ.png" /><figcaption>SOC workflow</figcaption></figure><p>Without that connection, a SOC can identify suspicious activity without having a consistent process for deciding what to do next.</p><p><strong>In this lab:</strong> Each playbook covers a group of related detections, not a single rule. For example, IR-002 (Active Directory Account &amp; Privilege Compromise) covers DET-007 to DET-016, so a new account, a password reset, and a Domain Admins change can be handled as one incident.</p><p><strong>8. Controlled Testing Exposes Detection Gaps</strong></p><p>Finally, controlled attack simulation was one of the most useful parts of the project.</p><p>Generating known activity and then tracing it through the environment allowed me to compare expected behavior with actual telemetry and detection results.</p><p>This exposed gaps that would not have been obvious from simply looking at the rule configuration.</p><p><strong>In this lab:</strong> Writing out DET-011 for this article showed its limits. It matches Event 4728 and the “Domain Admins” group only. Additions through local or universal groups (Events 4732 and 4756) need separate rules, and a nested group would not be caught at all. Tuning ideas such as covering other Tier-0 groups and raising severity for newly created accounts are the next step.</p><p>It also reinforced an important principle that carried throughout the project:</p><blockquote>A detection should be tested against the behavior it is designed to detect, not just checked for successful rule deployment.</blockquote><h4>What Building the Lab Changed for Me</h4><p>The biggest takeaway was that I stopped thinking about detection engineering as simply writing rules.</p><p>The process became much broader:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gBREO36CFT_1QTYyz4YPLQ.png" /><figcaption>Detection engineering lifecycle</figcaption></figure><p>Building the lab gave me practical experience with how these pieces depend on one another.</p><ul><li>A detection needs reliable telemetry.</li><li>An alert needs investigation context.</li><li>An investigation needs visibility.</li><li>A confirmed incident needs a response process.</li><li>And the results of that process can feed back into better detection.</li></ul><p>That end-to-end perspective is probably the most valuable lesson I took away from building this project.</p><h3>Inside the Repository</h3><p>The project was documented alongside the lab rather than treating documentation as an afterthought.</p><p>I organized the repository so that the detection rules, supporting documentation, incident response procedures, diagrams, and validation evidence could be explored independently.</p><p>The high-level structure looks like this:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/93080490ab2e1940fcf9e63b00da0f63/href">https://medium.com/media/93080490ab2e1940fcf9e63b00da0f63/href</a></iframe><p><strong>detection-rules/</strong></p><p>Contains the custom Wazuh detection rules developed for the project, along with documentation explaining their purpose and usage.</p><p><strong>docs/</strong></p><p>Contains the main technical documentation covering areas such as architecture, detection engineering, dashboards, validation, MITRE ATT&amp;CK, attack simulations, and lessons learned.</p><p><strong>docs/detections/</strong></p><p>Contains individual documentation for the 42 custom detections, organized by detection category.</p><p><strong>docs/mappings/</strong></p><p>Contains the relationships between detections and other parts of the security workflow, including:</p><ul><li>Detection → MITRE ATT&amp;CK</li><li>Detection → Windows Event ID</li><li>Detection → Incident Response Playbook</li><li>Detection Coverage Matrix</li></ul><p><strong>incident-response/</strong></p><p>Contains the seven incident response playbooks (IR-001 to IR-007): Credential Attack &amp; Account Compromise, Active Directory Account &amp; Privilege Compromise, Kerberos &amp; Credential Theft Attack, Active Directory Discovery &amp; Reconnaissance, Lateral Movement &amp; Remote Execution, Persistence &amp; Defense Evasion, and AD Destructive &amp; Impact Activity.</p><p><strong>diagrams/</strong></p><p>Contains the visual documentation for the project, including the lab architecture, telemetry flows, MITRE ATT&amp;CK coverage, attack-chain simulations, and the Detection → Investigation → Response workflow.</p><p><strong>screenshots/</strong></p><p>Contains dashboard recordings, screenshots, and other visual evidence from the lab, including the SOC dashboards and detection validation activities.</p><h4>A Documented SOC Lab</h4><p>Rather than keeping the detection rules, testing results, diagrams, and response procedures separate from the implementation, I wanted the repository to provide a complete view of how the SOC lab was built and operated.</p><h3>What This Lab Demonstrates</h3><p>By the end of the project, the lab had evolved into a small but complete security operations environment covering the core stages of the detection and response lifecycle.</p><p>The project includes:</p><ul><li><strong>42 custom Wazuh detections</strong> covering authentication, account management, privilege escalation, credential access, discovery, lateral movement, persistence, defense evasion, and impact.</li><li><strong>7 SOC dashboards</strong> providing visibility into authentication activity, Active Directory security, endpoint activity, threat hunting, MITRE ATT&amp;CK activity, and detection validation.</li><li><strong>7 incident response playbooks</strong> connecting security detections to structured investigation and response procedures.</li><li><strong>Windows and Sysmon telemetry</strong> collected from the Active Directory environment and monitored through Wazuh.</li><li><strong>MITRE ATT&amp;CK mapping</strong> connecting detected behaviors to established adversary techniques.</li><li>Controlled attack simulations used to generate security activity in the isolated lab.</li><li>A threat hunting dashboard for searching the telemetry for PowerShell activity, discovery tools, persistence, and lateral movement that may not have raised a dedicated alert.</li><li><strong>Detection validation</strong> performed by testing the detections against controlled activity and verifying the resulting telemetry and alerts.</li><li><strong>Incident investigation workflows</strong> used to move from an individual alert toward understanding the broader context of an event.</li><li><strong>Response workflows</strong> designed to connect investigation findings with containment, recovery, and follow-up actions.</li></ul><h4>From Individual Detections to SOC Operations</h4><p>The main outcome of the project was not simply the number of detection rules created.</p><p>It was the ability to connect the different components into an operational workflow:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BTTRzuFfefuZJFTkHrxZMA.png" /><figcaption>SOC project lifecycle</figcaption></figure><p>This helped demonstrate how telemetry, detection engineering, dashboards, threat hunting, incident investigation, and response procedures can work together as part of a SOC environment.</p><p>The project therefore represents more than a collection of Wazuh rules. It demonstrates a practical approach to building, testing, documenting, and operating a security monitoring and detection environment.</p><h3>Explore the Project</h3><p>The complete project is available on GitHub, including the detection rules, technical documentation, dashboards, attack simulations, MITRE ATT&amp;CK mappings, incident response playbooks, diagrams, screenshots, and detection validation documentation.</p><p>The repository is structured so that individual parts of the lab can be explored independently, while also showing how they fit together as an end-to-end SOC environment.</p><h4><strong>What You Can Explore</strong></h4><ul><li><strong>Detection Rules</strong> — 42 custom Wazuh detection rules and supporting documentation</li><li><strong>Architecture</strong> — Lab architecture, network design, and telemetry flows</li><li><strong>Dashboards</strong> — SOC monitoring, investigation, threat hunting, and detection validation dashboards</li><li><strong>Attack Simulations</strong> — Controlled activities used to test and validate detections</li><li><strong>MITRE ATT&amp;CK Mappings</strong> — Connections between detections, techniques, and security telemetry</li><li><strong>Incident Response Playbooks</strong> — Seven documented response procedures</li><li><strong>Detection Validation</strong> — Testing methodology and validation documentation</li><li><strong>Diagrams &amp; Screenshots</strong> — Visual documentation of the lab and its SOC workflows</li></ul><h4><strong>GitHub Repository</strong></h4><p>The complete project can be explored here:</p><p><a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab">SOC Active Directory Detection and Response Lab on GitHub</a></p><p>The repository contains the implementation and supporting evidence behind the workflows discussed throughout this article.</p><h3>Conclusion</h3><p>Building this project gave me a much deeper understanding of how the different parts of a SOC fit together, particularly within a Windows Active Directory environment.</p><p>What started as an Active Directory lab developed into a complete detection and response environment covering Windows and Sysmon telemetry, detection engineering, threat hunting, investigation, validation, dashboards, and incident response.</p><p>This two-part write-up documented both sides of the project. Part 1 focused on building the environment, engineering 42 custom Wazuh detections, mapping activity to MITRE ATT&amp;CK, simulating attacks, and creating SOC dashboards. Part 2 focused on detection validation, investigation, incident response, lessons learned, and connecting the different SOC workflows.</p><p><strong>Read Part 1:</strong> <a href="https://medium.com/@harikrishnan.p097/building-an-active-directory-soc-lab-detection-engineering-threat-hunting-incident-response-6f0f6c10cafd">Building an Active Directory SOC Lab: Detection Engineering, Threat Hunting &amp; Incident Response with Wazuh</a></p><p>The full project, including the rules, mappings, playbooks, and validation evidence, is in the <a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab">GitHub repository</a>.</p><p>One of the most valuable outcomes was the hands-on experience gained with Windows Active Directory infrastructure and AD-focused attack activity. Working with authentication events, account and group management, Kerberos activity, privilege changes, credential attacks, discovery, lateral movement, and other AD-related behaviors helped me understand how attacks can appear across enterprise Windows environments and how those behaviors can be detected through security telemetry.</p><p>This experience is particularly relevant to modern SOC operations, where Active Directory and Windows infrastructure remain critical components of many enterprise environments.</p><p>The biggest lesson I took from the project is that effective detection engineering is not simply about writing rules. It requires reliable telemetry, meaningful detection logic, validation, investigation context, security visibility, threat hunting, and defined response processes.</p><h3>Security Disclaimer</h3><p>This project was developed and tested in an isolated, controlled Active Directory lab environment for educational, research, and cybersecurity skill-development purposes.</p><p>The attack simulations, detection tests, and security techniques discussed in this series were performed only against systems that I controlled and had authorization to test. The examples are intended to demonstrate defensive security concepts, detection engineering, threat hunting, investigation, and incident response.</p><p>The techniques discussed should not be used against systems, networks, accounts, or applications without explicit authorization. Always perform security testing within an appropriately isolated and authorized environment.</p><h3>Connect With Me</h3><p>If you found this project interesting or would like to explore the implementation in more detail, you can find me here:</p><ul><li><strong>GitHub:</strong> <a href="https://github.com/Harikrishnan-P97">Harikrishnan-P97</a></li><li><strong>LinkedIn:</strong> <a href="http://www.linkedin.com/in/harikrishnanp097">Harikrishnan P</a></li><li><strong>Medium:</strong> <a href="https://medium.com/@harikrishnan.p097">@harikrishnan.p097</a></li></ul><p>I’m passionate about building practical cybersecurity projects and continuously developing my skills across SOC operations, detection engineering, blue-team security, Active Directory security, red teaming, and bug bounty research.</p><p>While this project focuses on the defensive side of security, exploring both offensive and defensive techniques has helped me better understand how attacks occur and how security teams can detect, investigate, and respond to them.</p><p>I’m continuing to build hands-on projects and labs to strengthen my understanding of real-world security operations and modern attack techniques.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd48c06fc5d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building an Active Directory SOC Lab: Detection Engineering, Threat Hunting & Incident Response…]]></title>
            <link>https://medium.com/@harikrishnan.p097/building-an-active-directory-soc-lab-detection-engineering-threat-hunting-incident-response-6f0f6c10cafd?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/6f0f6c10cafd</guid>
            <category><![CDATA[wazuh]]></category>
            <category><![CDATA[detection-engineering]]></category>
            <category><![CDATA[incident-response]]></category>
            <category><![CDATA[active-directory]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Fri, 02 Oct 2026 06:51:21 GMT</pubDate>
            <atom:updated>2026-10-04T08:52:58.006Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eBXCP1S1r6QktyBEm6dVEw.png" /></figure><h3>Building an Active Directory SOC Lab: Detection Engineering, Threat Hunting &amp; Incident Response with Wazuh</h3><h4>How I built, tested, and documented a hands-on security operations lab with 42 custom detections, Windows/Sysmon telemetry, MITRE ATT&amp;CK mapping, attack simulations, SOC dashboards, and incident response playbooks.</h4><p><strong>At a glance:</strong> 42 custom detections · 8 categories · MITRE ATT&amp;CK technique mapping · 7 SOC dashboards · 7 incident response playbooks · 3 VMs + Wazuh<br> <strong>Repository :</strong> <a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab">https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab</a></p><h4><strong>In this article</strong></h4><ul><li>Lab architecture</li><li>Where the security data comes from</li><li>Building 42 custom Wazuh detections</li><li>Representative detection examples</li><li>Mapping detections to MITRE ATT&amp;CK</li><li>Simulating attacker behavior</li><li>Following an attack through the SOC</li><li>Turning security events into SOC visibility</li><li>Limitations</li></ul><h3>Introduction</h3><p>Security Operations Centers (SOCs) depend on more than simply collecting security logs. Analysts need to understand what normal activity looks like, identify suspicious behavior, investigate alerts, and respond to threats using reliable processes.</p><p>To build practical experience with that workflow, I created an Active Directory SOC lab designed to simulate a small enterprise security monitoring environment. The lab provides a controlled environment for observing security activity across Windows, Active Directory, and endpoint systems.</p><p>The lab uses Wazuh as the SIEM and detection platform, with Windows Server 2022 Active Directory, Windows 11, and Sysmon providing endpoint and security telemetry. A Kali Linux machine is used as the controlled attacker and attack-simulation system, generating security activity that can be observed and detected across the environment. I developed and validated 42 custom Wazuh detections across eight categories: authentication, account management, privilege escalation, credential access, discovery, lateral movement, persistence and defense evasion, and impact.</p><p>The project also incorporates MITRE ATT&amp;CK mapping, controlled attack simulations, threat hunting, seven SOC dashboards, and seven incident response playbooks to connect detection engineering with practical SOC operations.</p><p>The result is a small, end-to-end SOC environment capable of collecting telemetry, detecting suspicious activity, validating detections, visualizing security events, and guiding analysts through investigation and response.</p><p>In this article, I’ll walk through how I built the lab, engineered the detections, simulated attacks, validated the results, and connected security alerts to an end-to-end SOC workflow.</p><h3>The Goal: Build a Detection-Focused SOC Lab</h3><p>When I started this project, I wanted to build more than a lab that could generate security alerts.</p><p>I wanted to build a small SOC environment where I could experience the complete lifecycle of security monitoring — from the moment an activity occurs on an endpoint to the point where an analyst can investigate it and determine an appropriate response.</p><p>The core workflow I wanted to demonstrate was:</p><p><strong>Telemetry → Detection → Alert → Investigation → Validation → Response</strong></p><p>This meant each part of the lab had a specific purpose.</p><p>Telemetry provided the raw evidence of activity through Windows Security events and Sysmon. Detection engineering transformed that telemetry into custom Wazuh rules designed to identify suspicious behavior. Alerts provided the initial signal for investigation, while threat hunting and investigation helped determine what happened and whether the activity was meaningful.</p><p>I then focused on validation by generating controlled attack activity in the isolated lab and verifying that the expected telemetry and detections were produced. Finally, incident response playbooks connected detections to practical response procedures.</p><p>This turned the lab into an end-to-end SOC workflow rather than a collection of independently deployed security tools.</p><h3>Lab Architecture</h3><p>To build the SOC workflow, I created an isolated lab environment consisting of a Wazuh SIEM, an Active Directory domain controller, a Windows endpoint, and a Kali Linux machine used for controlled attack simulations.</p><p>Each system has a specific role within the environment:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/4b6679a731a0bdf8f8b8c5e51bb11395/href">https://medium.com/media/4b6679a731a0bdf8f8b8c5e51bb11395/href</a></iframe><h4>How the Environment Works</h4><p>The basic telemetry flow through the lab is:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8GV70Bb94rguSxOEox2u6w.png" /><figcaption>SOC telemetry flow</figcaption></figure><p>The Windows systems are the primary sources of security telemetry. DC01 generates Active Directory and Windows security events related to authentication, account management, privilege changes, and other domain activity. CLIENT01 provides endpoint telemetry, including Windows Security events and Sysmon process and system activity.</p><p>Wazuh agents installed on the Windows systems collect the relevant telemetry and forward it to the Wazuh Manager. The manager processes the incoming events and applies the custom detection rules I developed for the lab.</p><p>The Kali Linux system is used to generate controlled attack and reconnaissance activity within the isolated environment. Depending on the scenario, the simulated activity targets the Windows endpoint or Active Directory environment. This allows me to observe how attacker behavior appears in Windows and Sysmon telemetry and determine whether the corresponding Wazuh detection is triggered.</p><p>When a detection matches the incoming telemetry, Wazuh generates an alert that can be reviewed through the SOC dashboards and investigated using the associated detection documentation and incident response playbooks.</p><p>This architecture provides the foundation for the rest of the project:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*emIYV3aM2Z0wslmoa9kmoQ.png" /><figcaption>Telemetry to response pipeline</figcaption></figure><h4>Architecture Diagram</h4><p>The following diagram provides a visual overview of how the systems, telemetry sources, Wazuh components, and SOC workflow are connected.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MFW9LRTxHhYkqvLrFEnt4Q.png" /><figcaption>Architecture diagram</figcaption></figure><h3>Where the Security Data Comes From</h3><p>Before building a detection, it is important to understand where the security data comes from.</p><p>In a SOC, telemetry is the raw security information generated by systems, applications, users, and network activity. This telemetry provides the evidence that analysts and detection rules use to identify potentially suspicious behavior.</p><p>In this lab, the primary telemetry comes from Windows Security Events, Active Directory, and Sysmon.</p><h4>Windows Security Events</h4><p>Windows records many security-related activities in the Windows Event Log. These events can provide information about authentication, account changes, group membership, privilege-related activity, and other security events.</p><p>For example:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/494a13b22ff2015a2e1f5fb27838bf26/href">https://medium.com/media/494a13b22ff2015a2e1f5fb27838bf26/href</a></iframe><p>Related group-membership events include 4732 (member added to a local group) and 4756 (member added to a universal group). DET-011 matches 4728 only, so additions through those group types would need separate rules.</p><p>These events can be useful individually, but they become much more valuable when multiple events are correlated.</p><p>For example, a single failed logon may simply be a user entering an incorrect password. However, a large number of failed logons against an account or from the same source can provide stronger evidence of suspicious authentication activity.</p><h4>Active Directory Telemetry</h4><p>Because the lab is built around an Active Directory environment, domain activity is another important source of security telemetry.</p><p>Changes such as creating or deleting user accounts, modifying group membership, enabling privileged accounts, and requesting Kerberos tickets can leave observable events in the environment.</p><p>This telemetry allows detections to focus not only on endpoint activity, but also on changes occurring within the identity and domain infrastructure.</p><h4>Sysmon Telemetry</h4><p>While Windows Security Events provide valuable information about authentication and administrative activity, Sysmon adds deeper visibility into endpoint behavior.</p><p>Sysmon can provide telemetry about activities such as:</p><ul><li>Process creation</li><li>Process relationships</li><li>Network connections</li><li>File creation and modification</li><li>Registry activity</li><li>Driver and image loading</li><li>Process execution, including PowerShell activity</li></ul><p>This additional visibility is particularly useful when investigating what happened on an endpoint after a suspicious action has been identified.</p><p>For example, an authentication event may tell us that a user successfully logged in, while Sysmon telemetry can provide additional context about processes that subsequently executed on the endpoint.</p><h4>Bringing the Telemetry Together</h4><p>The strength of the lab comes from combining these different telemetry sources.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K_sYodVpZLf5-qNKczT8lQ.png" /><figcaption>Telemetry sources</figcaption></figure><p>Each telemetry source provides a different part of the story.</p><p>Authentication events help answer who authenticated and whether it succeeded.<br>Active Directory events help answer what changed within the domain.<br>Sysmon events help answer what happened on the endpoint.</p><p>Together, they provide the context needed to build more meaningful detections and investigate suspicious activity.</p><h4>Why Telemetry Matters</h4><p>A detection rule is only as useful as the telemetry available to support it.</p><p>If the required event is not generated, collected, or forwarded correctly, even a well-designed detection rule may never trigger.</p><p>That is why telemetry validation was an important part of this project. Before relying on a detection, I needed to verify that the underlying activity generated the expected Windows or Sysmon events, that the Wazuh agent collected them, and that the Wazuh Manager received the data correctly.</p><h4>Telemetry Flow</h4><p>The following diagrams show how Windows and Sysmon telemetry moves through the environment before reaching the detection layer.</p><p><strong>Windows and Sysmon Telemetry Data Flow</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sX5FEkcrRtB4vIdL0LouOQ.png" /><figcaption>Windows and Sysmon telemetry dataflow</figcaption></figure><p><strong>Overall Telemetry Data Flow</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3RG4sbIf8QU83TaGZcOfDw.png" /><figcaption>Overall telemetry dataflow</figcaption></figure><p>Working with Active Directory also made one thing clear: identity, authentication, privilege, and endpoint activity are closely connected during an attack. A single suspicious login can lead to account changes, privilege escalation, Kerberos activity, credential access, or lateral movement. Building detections around these relationships helped me understand how Active Directory security depends on connecting activity across identity, authentication, and endpoint telemetry rather than looking at individual Windows events in isolation.</p><h3>Building 42 Custom Wazuh Detections</h3><p>With the telemetry flowing into Wazuh, the next challenge was turning that raw security data into something useful for a SOC analyst.</p><p>I developed 42 custom Wazuh detections instead of relying only on default detection logic, tailoring them to the activity and telemetry present in the lab.</p><p>The objective was to build detections around specific security behaviors, with the quality and purpose of each rule taking priority over the number of rules. Each detection was designed around the telemetry required to observe that behavior and the context an analyst would need to investigate the resulting alert.</p><h4><strong>Why Custom Detection Rules?</strong></h4><p>Default SIEM rules can provide a useful starting point, but a detection-focused lab also needs rules tailored to its environment and objectives.</p><p>For this project, custom rules allowed me to:</p><ul><li>Detect specific Active Directory and Windows security behaviors.</li><li>Correlate relevant authentication and account activity.</li><li>Identify suspicious privilege and group changes.</li><li>Detect credential-access and Kerberos-related activity.</li><li>Monitor discovery and lateral-movement techniques.</li><li>Detect persistence and defense-evasion behavior.</li><li>Generate alerts that could be validated against controlled attack activity.</li><li>Map detections to relevant MITRE ATT&amp;CK techniques.</li></ul><h4>How the Detections Were Organized</h4><p>The 42 detections were organized into eight security categories:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1fe8e685bbfb26c1497c29cff3d7ef6e/href">https://medium.com/media/1fe8e685bbfb26c1497c29cff3d7ef6e/href</a></iframe><p>This structure makes the detection set easier to maintain and also provides a way to connect individual detections with broader attacker behavior.</p><p>For example, authentication detections can identify suspicious login activity, while later categories cover what an attacker might attempt after gaining access, such as privilege escalation, credential access, lateral movement, persistence, or impact.</p><p>The complete detection set is shown below to provide a detailed view of the security behaviors covered by the lab. Each detection is identified by its detection ID, name, category, and relevant security telemetry, making it possible to see how the individual rules contribute to the overall detection coverage.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/d980c357bd64edfe1a846a33043fb4ac/href">https://medium.com/media/d980c357bd64edfe1a846a33043fb4ac/href</a></iframe><h4>From Telemetry to Detection</h4><p>The basic detection-engineering process followed this pattern:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OqZDgECOOA7P3LyBJ_k6yw.png" /><figcaption>Detection engineering process</figcaption></figure><h4>Representative Detection Examples</h4><p>After showing the complete detection set above, I’ll highlight a few representative detections to explain how individual rules were designed and validated.</p><p><strong>Example 1: Windows Failed Authentication</strong></p><p><strong>What It Detects</strong></p><blockquote><strong><em>DET-001</em></strong><em> (base rule) · feeds DET-002 and DET-006 · MITRE ATT&amp;CK: T1110 (Brute Force), applied at the threshold rules</em></blockquote><p>This detection identifies failed Windows authentication attempts.</p><p>A single failed login can be completely legitimate, such as a user entering an incorrect password. However, repeated failures can provide useful evidence when investigating suspicious authentication activity.</p><p>DET-001 is a base rule that tags every failed Windows logon (Event ID 4625) so other detections can build on it. DET-002 and DET-006 add the thresholds for brute-force and same-source failures.</p><p><strong>Telemetry</strong></p><p>The primary telemetry for this detection comes from Windows Security Event ID 4625, which records failed logon attempts.</p><p>The event provides information that can help establish details such as the account involved, authentication type, and source information.</p><p><strong>Why It Matters</strong></p><p>Failed authentication activity can be an early indicator of password guessing, brute-force activity, or attempts to access an account using invalid credentials.</p><p>The detection therefore provides an initial signal that can be correlated with additional authentication events.</p><p><strong>How It Was Validated</strong></p><p>I generated controlled failed authentication activity within the isolated lab and verified the expected Windows Security event.</p><p>The event was then confirmed as being collected by the Wazuh agent and received by the Wazuh Manager, after which the corresponding custom detection generated the expected alert.</p><p><strong>Example 2: Privileged Group Modification</strong></p><p><strong>What It Detects</strong></p><blockquote><strong><em>DET-011</em></strong><em> · MITRE ATT&amp;CK: T1098 (Account Manipulation)</em></blockquote><p>This detection fires when a member is added to the Domain Admins group. DET-012 and DET-013 cover Enterprise Admins and local Administrators.</p><p><strong>Telemetry</strong></p><p>The detection uses Windows Event ID 4728, which records a member being added to a security-enabled global group.</p><p>Here is the rule behind DET-011:</p><pre>&lt;rule id=&quot;100110&quot;<br>      level=&quot;12&quot;&gt;<br><br>    &lt;if_sid&gt;60103&lt;/if_sid&gt;<br><br>    &lt;field name=&quot;win.system.eventID&quot;&gt;4728&lt;/field&gt;<br><br>    &lt;field name=&quot;win.eventdata.targetUserName&quot;&gt;Domain Admins&lt;/field&gt;<br><br>    &lt;description&gt;<br>        DET-011 Member Added to Domain Admins: $(win.eventdata.memberName)<br>    &lt;/description&gt;<br><br>    &lt;group&gt;<br>        custom_windows,<br>        privilege_escalation,<br>        domain_admins,<br>        attack.t1098,<br>    &lt;/group&gt;<br><br>    &lt;mitre&gt;<br>        &lt;id&gt;T1098&lt;/id&gt;<br>    &lt;/mitre&gt;<br><br>&lt;/rule&gt;</pre><p>The rule builds on Wazuh’s Windows audit-success rule (60103), matches Event ID 4728 where the target group is &quot;Domain Admins,&quot; and raises a level 12 alert showing which member was added. It carries the ATT&amp;CK tag T1098.</p><p><strong>Why It Matters</strong></p><p>Changes to privileged group membership can significantly alter the access available to an account.</p><p>From a SOC perspective, the important question is whether the change was expected, authorized, and consistent with normal administrative activity.</p><p><strong>How It Was Validated</strong></p><p>I added a test user to Domain Admins under controlled conditions and confirmed that Event ID 4728 reached Wazuh and triggered DET-011.</p><p><strong>Example 3: Kerberoasting Detection</strong></p><p><strong>What It Detects</strong></p><blockquote><strong><em>DET-017</em></strong><em> · MITRE ATT&amp;CK: T1558.003 (Kerberoasting)</em></blockquote><p>This detection identifies activity associated with Kerberos service-ticket requests that may indicate Kerberoasting behavior.</p><p>Kerberoasting involves requesting service tickets associated with accounts configured with Service Principal Names (SPNs), with the objective of obtaining ticket material that can potentially be subjected to offline password cracking.</p><p><strong>Telemetry</strong></p><p>The primary Windows telemetry for this detection is Event ID 4769, which records Kerberos service-ticket requests.</p><p>DET-017 is deliberately broad: it fires on Kerberos service-ticket requests (Event ID 4769) and flags them as possible Kerberoasting. That maximizes visibility in a small lab, but it would be far too noisy in a production domain. A production version would add filters such as RC4 encryption (0x17), excluding machine accounts and krbtgt, and a request-volume threshold.</p><p><strong>Why It Matters</strong></p><p>Kerberos is fundamental to Active Directory authentication, so unusual service-ticket request activity can provide valuable detection opportunities.</p><p>However, legitimate applications and users can also generate service-ticket requests. This is why the detection must be considered in context rather than treating every 4769 event as malicious.</p><p><strong>How It Was Validated</strong></p><p>I requested service tickets for an SPN-enabled account from KALI and confirmed that the resulting 4769 events reached Wazuh and triggered DET-017.</p><p><strong>Example 4: Security Event Log Clearing</strong></p><p><strong>What It Detects</strong></p><blockquote><strong><em>DET-037</em></strong><em> · MITRE ATT&amp;CK: T1070.001 (Clear Windows Event Logs)</em></blockquote><p>This detection identifies attempts to clear the Windows Security Event Log.</p><p><strong>Telemetry</strong></p><p>The detection relies on Event ID 1102, which Windows writes when the Security log is cleared.</p><p>This type of event is particularly useful because clearing logs can remove evidence that analysts may need during an investigation.</p><p><strong>Why It Matters</strong></p><p>Log clearing can be associated with attempts to reduce visibility or remove traces of activity after other actions have occurred.</p><p>For a SOC, detecting this behavior can therefore be important for both investigation and incident response.</p><p><strong>How It Was Validated</strong></p><p>I cleared the Security log on a lab host and confirmed that the event reached Wazuh and triggered DET-037.</p><h4>Detection Engineering as an Iterative Process</h4><p>Developing the 42 detections was not a one-time process.</p><p>Each detection went through testing and validation to determine whether the underlying telemetry was available, whether the rule matched the intended behavior, and whether the resulting alert provided useful information for investigation.</p><p>Where necessary, detection logic could be tuned to improve the relationship between visibility and alert quality.</p><p>The complete detection documentation, including the individual detection logic, telemetry requirements, MITRE ATT&amp;CK mappings, validation details, and response guidance, is available in the project repository: <a href="https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab">https://github.com/Harikrishnan-P97/SOC-Active-Directory-Detection-and-Response-Lab</a></p><h3>Mapping Detections to MITRE ATT&amp;CK</h3><p>After developing the detection rules, the next step was to understand what attacker behaviors those detections represented.</p><p>For this, I used the <strong>MITRE ATT&amp;CK framework</strong>.</p><p>MITRE ATT&amp;CK is a knowledge base that organizes documented adversary tactics and techniques based on real-world observations of how attackers operate. For a detection engineer, it provides a useful common language for describing what a detection is designed to identify.</p><p>Instead of treating each Wazuh rule as an isolated security alert, MITRE ATT&amp;CK mapping allows the detection set to be viewed in the context of a broader attack lifecycle.</p><h4>Why MITRE ATT&amp;CK Mapping Matters</h4><p>The mapping helps answer a simple but important question:</p><blockquote>What attacker behavior does this detection represent?</blockquote><p>For example, a detection for suspicious Kerberos service-ticket requests can be connected to Kerberoasting, which is represented by the MITRE ATT&amp;CK technique T1558.003 — Steal or Forge Kerberos Tickets: Kerberoasting.</p><p>The relationship can be represented as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TBDtsv_UBB_2qAvO7HH3iw.png" /><figcaption>Kerberoasting detection flow</figcaption></figure><p>This creates a connection between attacker behavior, observable telemetry, detection logic, and SOC operations.</p><h4>Organizing Detection Coverage</h4><p>The 42 custom detections in the lab cover multiple areas of the attack lifecycle, including:</p><ul><li>Authentication</li><li>Account management</li><li>Privilege escalation</li><li>Credential access and Kerberos</li><li>Discovery and reconnaissance</li><li>Lateral movement and remote execution</li><li>Persistence and defense evasion</li><li>Impact</li></ul><p>Mapping these detections to MITRE ATT&amp;CK techniques provides a structured way to understand where the lab has detection visibility and which attacker behaviors are being monitored.</p><p>It also makes the detection engineering process easier to communicate. Rather than simply saying that the lab contains 42 rules, the mapping shows what those rules are intended to detect from an adversary-behavior perspective.</p><h4>From Technique to Detection</h4><p>The overall relationship can be thought of as:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0kWjDf4-W-qP3hIqYjKSKw.png" /><figcaption>Technique-to-detection mapping</figcaption></figure><p>This approach helped ensure that the detection rules were tied to meaningful security behaviors instead of being driven solely by individual event IDs.</p><p>The MITRE ATT&amp;CK mappings also provide a useful foundation for evaluating detection coverage and identifying areas where additional telemetry or detection logic could be developed in the future.</p><h4>MITRE ATT&amp;CK Coverage</h4><p>The following diagram provides a visual overview of the ATT&amp;CK techniques covered by the project’s detection set.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*11-COrfo21FR2Iw0uc4uyg.png" /><figcaption>MITRE ATT&amp;CK coverage matrix</figcaption></figure><h3>Simulating Attacker Behavior</h3><p>Creating a detection rule is only the beginning. To determine whether a detection actually works, I needed to generate the behavior it was designed to identify and verify the entire telemetry and alerting path.</p><p>For this project, all attack simulations were performed in an isolated lab environment. The objective was to generate controlled security activity, observe the resulting telemetry, and determine whether the expected Wazuh detection was triggered.</p><p>Depending on the technique being tested, activity was generated from KALI, CLIENT01, or DC01.</p><p><strong>Validation Methodology</strong></p><p>Each detection followed a structured validation process:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I3aPp299cNHZ2d4uOfR0tA.png" /><figcaption>Validation methodology</figcaption></figure><p>The first step was to select the detection being tested and identify the corresponding attacker behavior or MITRE ATT&amp;CK technique.</p><p>I then generated the relevant activity within the isolated environment. For example, authentication detections could be tested by generating controlled failed or successful logon activity, while other detections required account changes, privilege modifications, discovery activity, remote execution, persistence techniques, or defense-evasion behavior.</p><p><strong>Verifying the Telemetry</strong></p><p>Generating the activity was not enough.</p><p>I first verified that the expected Windows Security Event or Sysmon event was actually produced. This helped confirm that the underlying operating system telemetry contained the evidence required by the detection.</p><p>The telemetry was then checked to ensure that it was collected by the appropriate Wazuh agent and successfully forwarded to the Wazuh Manager.</p><p>This step was important because a detection can fail for reasons unrelated to the rule itself. If the required telemetry is not generated or collected correctly, the detection cannot operate as intended.</p><p><strong>Verifying the Wazuh Alert</strong></p><p>After confirming the telemetry, I verified whether the corresponding custom Wazuh rule matched the event and generated the expected alert.</p><p>This validated the complete telemetry-to-alert path:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qo9UeSR_QfNj4ULqJUOOqg.png" /><figcaption>Wazuh alert verification</figcaption></figure><p>If the expected alert was generated, I could then examine the alert details and confirm that the detection was identifying the intended behavior.</p><p><strong>Cleanup and Documentation</strong></p><p>After each simulation, cleanup was performed where applicable to return the environment to an appropriate state for subsequent testing.</p><p>The results of the validation were then documented, including the activity generated, telemetry observed, detection triggered, and relevant investigation or response information.</p><p>I repeated this process across the detection set.</p><p><strong>From Individual Tests to an Attack Chain</strong></p><p>Testing individual detections is useful, but security incidents rarely consist of a single isolated event.</p><p>The project therefore also used controlled attack sequences to demonstrate how multiple activities could appear across the environment and how individual detections could contribute to a broader investigation.</p><p>The following diagram illustrates an example attack chain and how activity can progress through telemetry, detection, and SOC analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UoEADZ1e83qKI3iCLLyE9A.png" /><figcaption>Sample attack chain simulation</figcaption></figure><p>The key objective was to verify the complete detection pipeline, from the original activity through telemetry generation, detection, alerting, and validation:</p><blockquote><em>Generate the behavior → observe the telemetry → trigger the detection → verify the alert → validate the result → document the outcome.</em></blockquote><p>This provided a repeatable way to evaluate whether the detection engineering work translated into observable and actionable SOC alerts.</p><h3>Following an Attack Through the SOC</h3><p>So far, I have described the lab architecture, telemetry sources, detection engineering process, MITRE ATT&amp;CK mapping, and attack-simulation methodology.</p><p>But what does all of this actually look like during a security investigation?</p><p>To make the workflow easier to understand, it helps to follow a single activity from the moment it occurs to the point where an analyst can investigate and respond to it.</p><p>For this example, consider a controlled authentication scenario in which multiple failed authentication attempts are followed by a successful login.</p><p>The overall flow looks like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mxdMX4wlrzo8POAQjIpNzQ.png" /><figcaption>Authentication attack flow</figcaption></figure><p><strong>1. Attack Activity</strong></p><p>The activity begins with controlled authentication attempts against a Windows system in the isolated lab.</p><p>Multiple failed authentication attempts are generated, followed by a successful authentication.</p><p>From an analyst’s perspective, the important part is the context surrounding the login. The sequence of events can reveal additional information that helps determine whether the activity warrants further investigation.</p><p><strong>2. Windows Telemetry</strong></p><p>The authentication activity generates Windows Security events.</p><p>For example, failed authentication attempts generate Event ID 4625, while a successful logon generates Event ID 4624.</p><p>These events provide information that can help establish details such as the account involved, the authentication activity, and the source of the request.</p><p>At this point, the activity exists as raw telemetry rather than a SOC alert.</p><p><strong>3. Wazuh Collection</strong></p><p>The Wazuh agent on the Windows system collects the relevant security telemetry and forwards it to the Wazuh Manager.</p><p>The manager can then process the events against the custom detection rules developed for the lab.</p><p>This is where raw Windows telemetry starts becoming security-relevant information.</p><p><strong>4. Detection</strong></p><p>The relevant custom Wazuh detection evaluates the authentication activity.</p><p>In this scenario, the sequence of failed authentication attempts followed by a successful login can provide the context required for DET-004 — Successful Login After Failed Attempts.</p><p>Instead of looking at the successful login in isolation, the detection considers the preceding failed authentication activity.</p><p><strong>5. Alert</strong></p><p>When the detection conditions are satisfied, Wazuh generates an alert.</p><p>The alert gives the analyst a starting point for investigation and can contain information such as the affected account, source information, event details, timestamps, and detection metadata.</p><p>The important transition here is:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QyIoIwmisfbtMXs4R__aqQ.png" /><figcaption>Wazuh alert processing flow</figcaption></figure><p><strong>6. Dashboard</strong></p><p>The alert becomes visible within the SOC monitoring environment through the Wazuh dashboards.</p><p>The analyst can use the dashboard to understand the broader context of the activity, including alert severity, affected users or hosts, detection categories, and related events.</p><p>This is where individual alerts become part of the wider SOC monitoring workflow.</p><p><strong>7. Investigation</strong></p><p>The analyst can then investigate the alert by asking questions such as:</p><ul><li>Which account was involved?</li><li>Where did the authentication attempts originate?</li><li>How many failed attempts occurred?</li><li>When did the successful authentication take place?</li><li>Were there other suspicious events around the same time?</li><li>Was the activity expected or authorized?</li><li>Did additional activity occur after the successful login?</li></ul><p>The goal is to move beyond the initial alert and establish what actually happened.</p><p>This is where correlation between Windows Security telemetry, Sysmon activity, user information, source information, and other related alerts can provide additional context.</p><p><strong>8. Response</strong></p><p>If the investigation determines that the activity represents a security incident, the analyst can follow the appropriate incident response procedure.</p><p>The response may involve actions such as containing the affected account or endpoint, reviewing related activity, collecting additional evidence, and documenting the incident.</p><p>The specific response depends on the nature and severity of the activity rather than treating every alert in exactly the same way.</p><h3>The SOC Workflow in Practice</h3><p>This example demonstrates the relationship between the different components of the project:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vUR7uRJFyMgaqbHThKByDQ.png" /><figcaption>SOC workflow</figcaption></figure><p>For a beginner, this is the most important concept to take away from the project:</p><p>A SOC is not just a dashboard full of alerts.</p><p>The value comes from connecting telemetry, detection logic, investigation, and response into a repeatable workflow.</p><p>That connection is what turns individual security events into actionable security operations.</p><h3>Turning Security Events into SOC Visibility</h3><p>A detection is useful only if an analyst can quickly understand what is happening around it.</p><p>To make the lab’s security activity easier to monitor and investigate, I built seven SOC dashboards in Wazuh. Each one is designed around a specific question an analyst might need to answer during monitoring or investigation, so the dashboards support analysis rather than just visualization.</p><p><strong>SOC Dashboards</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fZsRCkCNSxIBQrAqVchYwg.png" /><figcaption>Wazuh SOC dashboards</figcaption></figure><p><strong>1. SOC Detection Overview</strong></p><p><strong>Analyst question:</strong> <em>What security activity is happening across the environment?</em></p><p>This dashboard provides a high-level view of detection activity, including alert volume, severity, detection categories, affected hosts, affected users, and recent high-severity alerts.</p><p>It serves as the starting point for understanding the overall security posture of the lab and identifying areas that may require further investigation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xaog17Gdn0WIHVac_5eoNA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*i0AfPNZQAb_ffV6WSV_UhA.png" /><figcaption>SOC detection overview</figcaption></figure><p><strong>2. Authentication &amp; Account Monitoring</strong></p><p><strong>Analyst question:</strong> <em>Who is authenticating, and where are authentication failures occurring?</em></p><p>This dashboard focuses on successful and failed logons, account lockouts, privileged logons, affected users, and source information.</p><p>It can help an analyst investigate unusual authentication patterns, repeated failures, or activity involving specific accounts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Vc-Az3_djSeGJwvFePcFkA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4T41AxkNdlOedixt-MB91Q.png" /><figcaption>Authentication and account monitoring</figcaption></figure><p><strong>3. Active Directory Security</strong></p><p><strong>Analyst question:</strong> <em>What is changing inside Active Directory?</em></p><p>This dashboard focuses on domain-level security activity such as account changes, group membership modifications, privileged account activity, and other Active Directory-related events.</p><p>For an analyst, this provides visibility into changes that could affect identity and access within the domain.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZZSGIha4Y_vTic6IBiR3_w.png" /><figcaption>Active Directory security</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t3EPn-qyd0Kr4la_Q6COdw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RSpDbqQxvJy8dotC7bDssQ.png" /><figcaption>Active Directory security</figcaption></figure><p><strong>4. Sysmon Endpoint Activity</strong></p><p><strong>Analyst question:</strong> <em>What is happening on the Windows endpoints?</em></p><p>This dashboard provides visibility into endpoint activity collected through Sysmon.</p><p>Process execution, network activity, and other endpoint telemetry can provide additional context when investigating a suspicious event.</p><p>For example, an authentication alert may tell an analyst that an account logged in, while endpoint telemetry can help investigate what happened on the system after that authentication.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UMwUxeGWBj1BdV8oZNy7Lg.png" /><figcaption>Sysmon endpoint activity</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*f0yvYWGlE2As580fDMyNvw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6S5_p8X22NjXH-2eOKaYSA.png" /><figcaption>Sysmon endpoint activity</figcaption></figure><p><strong>5. Threat Hunting &amp; Lateral Movement</strong></p><p><strong>Analyst question:</strong> <em>Is there evidence of suspicious movement between systems?</em></p><p>This dashboard focuses on activity that can help identify potential reconnaissance, remote execution, and lateral-movement behavior.</p><p>It provides an investigation-oriented view where related activity can be examined across systems, giving analysts broader context than a single alert can provide.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eKJX0ILQnE1VDwY_Rv-6LA.png" /><figcaption>Threat hunting and lateral movement</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4wuIR3EjFIKUyn2_jx8alQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rANUWRkxiQ-YJ-eR3-4c6w.png" /><figcaption>Threat hunting and lateral movement</figcaption></figure><p><strong>6. MITRE ATT&amp;CK Coverage &amp; Activity</strong></p><p><strong>Analyst question:</strong> <em>What attacker techniques are being detected?</em></p><p>This dashboard connects detection activity with the MITRE ATT&amp;CK framework.</p><p>It provides a way to visualize which documented attacker techniques are represented within the detection set and how those techniques appear in observed security activity.</p><p>This connects the detection engineering work described earlier in the article with the actual security activity being generated in the lab.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tk9CR58JM4-OEWLF6Ed3Bw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pow0hiUbWTHyw3KyNpSJkg.png" /><figcaption>MITRE ATT&amp;CK coverage and activity</figcaption></figure><p><strong>7. Detection Validation &amp; Engineering</strong></p><p><strong>Analyst question:</strong> <em>Are the detections working as expected?</em></p><p>This dashboard focuses on the detection-engineering side of the environment.</p><p>It helps review detection activity, alert generation, and validation results during controlled testing. This is particularly useful when developing or tuning rules because a detection needs to be tested against the telemetry it is designed to identify.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*llnvjgR83oQrJ6WmQOFmvg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0kCRzdcm0XIk2XXc0eQ3ZA.png" /><figcaption>Detection validation and engineering</figcaption></figure><p>Together, these dashboards turn the large volume of Windows, Active Directory, and Sysmon telemetry generated by the lab into a more structured view of security activity.</p><p><strong>From Alerts to Investigation</strong></p><p>The dashboards are not intended to replace the underlying alerts or event data.</p><p>Instead, they provide different levels of visibility:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3o3zi0HMHa-VLycrp0zeiA.png" /><figcaption>Alert to investigation workflow</figcaption></figure><p>For example, an analyst might begin with a spike in failed authentication activity on the SOC Detection Overview dashboard, move to Authentication &amp; Account Monitoring to identify affected accounts and sources, and then use Sysmon Endpoint Activity to investigate what occurred on the relevant endpoint.</p><p>This makes the dashboards part of the investigation workflow instead of treating them as a collection of charts.</p><h3>Limitations</h3><p>This is a small, isolated lab with simulated attacks and no real user noise. The detections are validated for coverage, not tuned for production false-positive rates, and several rules would need baselining and allow-listing in a real environment. Coverage is also uneven: Impact has a single detection (DET-042), while credential access and persistence/defense evasion are covered in more depth.</p><p>Some rules, such as DET-017 for Kerberoasting, are intentionally broad and would need filtering to be usable at scale.</p><h3>What&#39;s Next</h3><p>Detection and visibility are only the starting point. In Part 2, I move into detection validation results, investigation, incident response playbooks, and lessons learned.</p><p><a href="https://medium.com/@harikrishnan.p097/operating-an-active-directory-soc-lab-detection-validation-investigation-incident-response-dd48c06fc5d8?postPublishedType=initial">Part 2 — Operating an Active Directory SOC Lab: Detection Validation, Investigation &amp; Incident Response</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6f0f6c10cafd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Simulating a Real Security Operations Center: An End-to-End SOC Lab Project]]></title>
            <link>https://medium.com/@harikrishnan.p097/simulating-a-real-security-operations-center-an-end-to-end-soc-lab-project-41cd4542d78c?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/41cd4542d78c</guid>
            <category><![CDATA[blue-team]]></category>
            <category><![CDATA[homelab]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[detection-engineering]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Mon, 29 Jun 2026 08:29:31 GMT</pubDate>
            <atom:updated>2026-06-29T08:29:31.038Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aN0EciuzwaWCRfRcmyZuNw.png" /></figure><h4>Building a complete attack-to-response workflow that combines detection engineering, threat intelligence enrichment, analyst investigations, and automated incident response.</h4><h3>Introduction</h3><h4>Why I Built This Lab</h4><p>Over the past several months, I’ve been building home lab projects to better understand how cyber attacks unfold and how defenders detect and respond to them. Along the way, I explored topics such as the Cyber Kill Chain, phishing simulations, command-and-control infrastructure, and detection engineering. Each project taught me something new, but they also revealed an important gap in my understanding.</p><p>Most security labs focus on individual technologies or isolated attack scenarios. You might learn how to configure a SIEM, write a detection rule, or deploy an endpoint sensor, but real Security Operations Centers (SOCs) don’t operate as disconnected tools. Every alert passes through a much larger workflow involving data collection, investigation, enrichment, analyst decision-making, and incident response.</p><p>That realization became the motivation for this project.</p><p>Rather than building another lab around a single product or detection technique, I wanted to simulate the complete lifecycle of a security incident — from the moment malicious activity occurs to the point where an analyst investigates the alert, enriches it with threat intelligence, and initiates response actions.</p><p>The result is an end-to-end SOC automation and detection engineering lab that combines telemetry collection, SIEM detections, SOAR automation, investigation dashboards, threat intelligence enrichment, and automated containment into a single integrated environment.</p><p>The objective wasn’t simply to learn individual tools. It was to understand how those tools work together to support the daily operations of a modern Security Operations Center.</p><h4>What Readers Will Learn</h4><p>In this article, I’ll walk through the architecture and design of the entire environment, explaining how attack telemetry flows through the lab, how detection rules generate meaningful alerts, how automation streamlines investigations, and how analysts can progress from initial detection to containment using a unified workflow.</p><p>We’ll explore how the different components interact, why each technology was selected, and the lessons learned while building an environment that closely resembles the workflows used by real SOC teams.</p><h4>Previous Articles in This Series</h4><p>This project builds upon two earlier labs that explored the offensive and defensive aspects of phishing simulations:</p><ul><li><a href="https://medium.com/@harikrishnan.p097/i-built-a-phishing-lab-to-finally-understand-the-cyber-kill-chain-9aad7c10e8c4"><strong>I Build a Phishing Lab to Finally Understand the Cyber Kill Chain</strong></a></li><li><a href="https://medium.com/@harikrishnan.p097/inside-my-phishing-simulation-lab-red-team-meets-blue-team-05731f401bb9"><strong>Inside My Phishing Simulation Lab: Red Team Meets Blue Team</strong></a></li></ul><p>Those articles focused primarily on attack simulation, phishing infrastructure, and telemetry generation. This article continues that journey by showing what happens after the telemetry reaches a Security Operations Center — how events are detected, investigated, enriched with threat intelligence, and ultimately transformed into actionable incident response.</p><h3>Understanding the Challenge</h3><h4>Why Traditional Labs Fall Short</h4><p>Most cybersecurity labs are designed to teach individual technologies or techniques. You might build a SIEM lab to learn how to write detection rules, deploy an endpoint monitoring solution to collect telemetry, or experiment with attack simulations to understand adversary behavior. While each of these exercises is valuable, they often exist in isolation.</p><p>In a real Security Operations Center, security tools rarely operate independently. An alert generated by a detection rule is only the beginning of a much larger process. Analysts must determine whether the alert is legitimate, gather additional context, correlate related events, assess the severity of the activity, and decide whether further action is required. Each step depends on information collected from multiple systems working together.</p><p>This was the biggest limitation I encountered while building previous labs. I could generate telemetry, create detection rules, and trigger alerts, but there was very little context around what happened after an alert appeared. There was no structured investigation workflow, no automated enrichment, no analyst case management, and no integrated response process.</p><p>The objective of this project was to bridge that gap by moving beyond isolated demonstrations and building an environment that reflects how a modern SOC operates. Instead of treating detection engineering, automation, investigation, and response as separate topics, I wanted to connect them into a single end-to-end workflow.</p><h4>The Gap Between Alerts and Incident Response</h4><p>Generating alerts is relatively easy. Turning those alerts into meaningful investigations is much more difficult.</p><p>A busy SOC may receive hundreds or even thousands of alerts every day. Without sufficient context, analysts can spend significant amounts of time manually collecting information before they are able to determine whether an alert represents a genuine threat or simply expected system behavior.</p><p>This challenge is often referred to as <strong>alert fatigue</strong>. Analysts are required to review large numbers of alerts, many of which may be low priority, duplicates, or false positives. When every alert requires manual investigation, response times increase and important incidents can be overlooked.</p><p>Another challenge is the lack of enrichment. A detection rule might identify a suspicious IP address or file hash, but it rarely provides enough information for an analyst to immediately understand its significance. External threat intelligence, asset context, historical activity, and investigation timelines all contribute to a more complete understanding of an incident.</p><p>Finally, there is the issue of response. Detecting malicious activity has little operational value if analysts cannot quickly contain the affected systems. Modern SOCs increasingly rely on automation to perform repetitive investigation tasks, enrich alerts with additional context, notify analysts, and initiate response actions such as endpoint isolation or case escalation.</p><p>These challenges shaped the design of this lab. Rather than focusing solely on generating alerts, the environment was built to simulate the complete operational lifecycle of a security incident — from telemetry collection and detection engineering through investigation, enrichment, analyst decision-making, and automated response.</p><h3>Designing the SOC Architecture</h3><h4>High-Level Architecture Overview</h4><p>Once I had identified the limitations of traditional cybersecurity labs, the next challenge was designing an environment that could simulate the complete lifecycle of a security incident.</p><p>Rather than treating security monitoring as a collection of independent tools, I wanted every component to contribute to a single operational workflow. An attack should generate telemetry, telemetry should feed the SIEM, detections should trigger automation, automation should enrich alerts, analysts should investigate those alerts through dedicated dashboards, and response actions should ultimately contain the incident.</p><p>This design philosophy became the foundation of the lab.</p><p>Instead of optimizing for complexity, I focused on creating clear data flows between components so that every stage of the SOC lifecycle could be observed, understood, and improved independently.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nPvzjwu_x8lXRm5ZZP4zIw.png" /><figcaption>Figure 1. High-level architecture of the SOC Automation &amp; Detection Engineering Lab.</figcaption></figure><p>At a high level, the environment is divided into four logical layers:</p><ul><li><strong>Attack Simulation</strong> — Generates realistic attacker activity, including phishing campaigns, endpoint execution, network reconnaissance, web application attacks, and data exfiltration.</li><li><strong>Monitoring Layer</strong> — Collects telemetry from multiple sources including Windows Event Logs, Sysmon, Suricata IDS, ModSecurity WAF, and web application logs.</li><li><strong>SIEM &amp; Automation Layer</strong> — Splunk performs detection engineering and alert generation, while n8n automates alert enrichment, case management, dashboard APIs, notifications, and response workflows.</li><li><strong>SOC Operations &amp; Response Layer</strong> — Grafana provides analyst dashboards and investigation workflows, VirusTotal supplies external threat intelligence enrichment, LimaCharlie enables endpoint response actions, and Slack delivers operational notifications.</li></ul><p>Together, these layers simulate how information moves through a modern Security Operations Center, transforming raw telemetry into actionable security incidents.</p><h3>Core Design Goals</h3><p>Before selecting specific technologies, I established several design principles that guided the architecture of the lab.</p><h4>End-to-End Visibility</h4><p>One of the primary objectives was to ensure complete visibility across the incident lifecycle. Every stage — from attack execution and telemetry generation to detection, investigation, enrichment, and containment — needed to be observable.</p><p>Rather than focusing solely on alerts, I wanted to understand how events evolve as they move through the SOC.</p><h4>Realistic Analyst Workflows</h4><p>Many cybersecurity labs stop once an alert appears in a SIEM.</p><p>In reality, analysts spend far more time investigating alerts than generating them. The environment therefore includes investigation dashboards, alert timelines, asset context, analyst notes, escalation workflows, and case status tracking to better reflect day-to-day SOC operations.</p><h4>Automated Enrichment</h4><p>Alerts rarely contain enough information to support immediate decision-making.</p><p>The architecture automatically enriches indicators such as IP addresses, domains, and file hashes using VirusTotal, allowing analysts to quickly understand the reputation of observed artifacts without leaving the investigation workflow.</p><h4>Response Orchestration</h4><p>Detection has limited value if analysts cannot rapidly respond.</p><p>The lab therefore integrates automated response capabilities, enabling workflows such as Slack notifications, case updates, and endpoint isolation through LimaCharlie directly from investigation dashboards.</p><h4>Scalability and Modularity</h4><p>Finally, every major component was designed to operate as an independent module.</p><p>Detection rules can be expanded without changing automation workflows. New telemetry sources can be added without redesigning dashboards. Additional enrichment providers can be integrated without affecting the investigation process.</p><p>This modular approach makes the environment easier to maintain while providing a foundation for future experimentation.</p><h3>Key Architectural Components</h3><p>Although the architecture contains several technologies, each one serves a clearly defined purpose within the SOC workflow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QcG9xwbNVQ5E8hUrE0yaMA.png" /><figcaption>Figure 2. SOC Lab Components Overview</figcaption></figure><p>Each technology addresses a specific stage of the incident lifecycle, but their real value comes from how they interact as part of a unified workflow rather than as standalone security products.</p><h3>Building the Lab Environment</h3><p>Designing the SOC architecture was only the first step. The next challenge was creating an environment capable of generating realistic attack telemetry while supporting the monitoring, investigation, and response workflows of a modern Security Operations Center.</p><p>To achieve this, I built a fully virtualized lab using Oracle VirtualBox. The environment consists of multiple virtual machines, each representing a specific role within the SOC ecosystem. Rather than relying on cloud-hosted services or isolated demonstrations, every major component was deployed within a controlled lab network, allowing attacks to be executed safely while producing realistic telemetry for analysis.</p><p>The result is a self-contained environment where offensive activities, security monitoring, analyst investigations, and automated response workflows can all be observed from beginning to end.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1sW6GTxy7UkO2NlfQVzb9A.png" /><figcaption>Figure 3. Virtualbox Lab Overview</figcaption></figure><h4>Virtualized Infrastructure</h4><p>The lab was designed around the principle that every component should contribute to the overall SOC workflow. Instead of deploying a single machine running multiple services, the infrastructure was divided into dedicated systems that closely resemble the separation of responsibilities found in production environments.</p><p>This approach makes it easier to understand how telemetry flows between systems while also allowing individual components to be expanded or replaced as the lab evolves.</p><h4>VirtualBox Environment</h4><p>Oracle VirtualBox serves as the foundation of the entire lab.</p><p>It hosts the Windows endpoints, security monitoring infrastructure, automation platform, and supporting services required for attack simulation and incident response. Using virtualization also makes the environment highly portable, allowing snapshots to be taken before experiments and quickly restoring systems after testing.</p><h4>pfSense Firewall</h4><p>pfSense acts as the central gateway for the virtual network.</p><p>In addition to providing routing and network segmentation, it allows traffic to flow between different parts of the lab while maintaining a realistic network topology. This separation makes it possible to simulate attacker movement, monitor network activity, and generate IDS telemetry through Suricata.</p><h4>Windows Endpoints</h4><p>The Windows virtual machines represent managed endpoints within the organization.</p><p>These systems generate the majority of the telemetry used throughout the project, including Windows Event Logs, Sysmon events, PowerShell activity, authentication events, process creation, registry modifications, and network connections.</p><p>They also serve as the targets for phishing campaigns, endpoint detections, and incident response actions.</p><h4>Splunk Server</h4><p>Splunk Enterprise acts as the central SIEM platform for the environment.</p><p>Telemetry collected from Windows endpoints, network sensors, and web application logs is ingested, normalized, and processed through custom correlation searches and detection rules. When suspicious activity is identified, alerts are generated and forwarded into the automation workflows.</p><h4>Grafana Server</h4><p>Grafana provides the operational interface used by SOC analysts.</p><p>Instead of interacting directly with raw log data, analysts use dashboards to prioritize alerts, investigate incidents, review timelines, monitor SLA metrics, and track the progress of investigations.</p><p>This layer transforms detection data into an analyst-friendly workflow.</p><h4>n8n Automation Server</h4><p>n8n functions as the orchestration engine for the lab.</p><p>Whenever Splunk generates an alert, n8n enriches the event, updates the SOC case lifecycle, communicates with Grafana APIs, integrates with VirusTotal, sends Slack notifications, and coordinates response workflows.</p><p>By automating repetitive tasks, the platform allows analysts to focus on investigation rather than manual data collection.</p><h4>LimaCharlie Integration</h4><p>LimaCharlie provides the endpoint response capabilities of the environment.</p><p>Once an analyst has validated an incident, response actions such as host isolation can be initiated directly from the investigation workflow. This closes the loop between detection and containment, demonstrating how modern SOCs combine monitoring with active response.</p><h4>Building a Modular Environment</h4><p>One of the primary design goals was modularity.</p><p>Every major component operates independently while communicating through clearly defined interfaces and APIs. This makes the lab easier to maintain, simplifies troubleshooting, and allows new telemetry sources, detection rules, automation workflows, or response capabilities to be added without redesigning the entire environment.</p><p>More importantly, this modular design closely reflects how enterprise SOCs evolve over time — new technologies can be introduced as requirements change without disrupting existing workflows.</p><h3>Building the Telemetry Pipeline</h3><p>A Security Operations Center is only as effective as the telemetry it receives. Regardless of how sophisticated detection rules or automation workflows may be, they depend entirely on the quality, coverage, and context of the underlying data.</p><p>For this reason, one of the primary goals of the lab was to build a telemetry pipeline capable of collecting events from multiple sources, normalizing them within the SIEM, and transforming them into actionable security alerts.</p><p>Rather than relying on a single log source, the environment combines endpoint, network, web application, and attack simulation telemetry to provide a comprehensive view of security events across the lab.</p><h3>Data Sources</h3><p>To simulate the diverse data available to a modern Security Operations Center, telemetry is collected from multiple layers of the environment.</p><h4>Windows Event Logs</h4><p>Windows Event Logs provide native operating system visibility, including authentication events, account management, privilege changes, and other security-relevant activities.</p><p>These events form the foundation for detections such as failed logons, successful logons, account lockouts, and privilege escalation attempts.</p><h4>Sysmon</h4><p>While Windows Event Logs provide basic security auditing, Sysmon significantly expands endpoint visibility.</p><p>It records detailed information about process creation, parent-child relationships, command-line arguments, network connections, registry modifications, driver loading, and file creation events.</p><p>This richer telemetry enables behavioral detections such as suspicious PowerShell execution, credential dumping, post-exploitation reconnaissance, and data exfiltration.</p><h4>Suricata</h4><p>Suricata provides network-based telemetry by monitoring traffic flowing through the virtual environment.</p><p>It generates alerts for reconnaissance activity, port scanning, brute-force attempts, and other suspicious network behavior that may not be visible from endpoint telemetry alone.</p><h4>ModSecurity</h4><p>To monitor attacks targeting web applications, the lab uses ModSecurity with the OWASP Core Rule Set (CRS).</p><p>HTTP requests are inspected for common attack techniques such as SQL injection, Cross-Site Scripting (XSS), directory traversal, and command injection, providing visibility into web-based attack activity.</p><h4>DVWA</h4><p>The Damn Vulnerable Web Application (DVWA) serves as the target application for web attack simulations.</p><p>Rather than existing solely as a vulnerable application, DVWA generates realistic HTTP traffic that allows ModSecurity and Splunk detections to be validated against genuine attack scenarios.</p><h4>Phishing Simulation Infrastructure</h4><p>The phishing environment acts as another source of telemetry within the SOC.</p><p>Email delivery, user interaction, payload execution, command-and-control communication, and simulated data exfiltration all generate events that propagate through the telemetry pipeline, allowing the complete attack lifecycle to be observed from initial access through incident response.</p><h3>Data Flow</h3><p>Once telemetry is generated, it follows a structured path through the SOC environment before reaching analysts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_QN2V5fGSaDxa1aH56fykA.png" /><figcaption>Figure 4. End-to-end telemetry pipeline from attack activity to incident response.</figcaption></figure><p>Each stage adds additional context to the original event. What begins as raw telemetry is gradually transformed into a structured incident that analysts can investigate and respond to efficiently.</p><h3>Why Telemetry Normalization Matters</h3><p>Collecting large volumes of telemetry is only part of the challenge. Different technologies produce events in different formats, with varying field names, timestamps, and metadata.</p><p>Without normalization, analysts would need to interpret each log source individually, making correlation difficult and increasing investigation time.</p><p>Splunk acts as the central aggregation point for the environment by ingesting telemetry from multiple sources and presenting it in a consistent format. This enables correlation searches to combine endpoint, network, and web application events into unified detections rather than isolated alerts.</p><p>Normalization also simplifies downstream automation. Because alerts follow a consistent structure, n8n workflows can enrich events, update case records, trigger notifications, and coordinate response actions without requiring separate logic for every telemetry source.</p><p>Ultimately, the objective is not simply to collect more logs — it is to transform diverse security events into meaningful, actionable information that supports rapid investigation and response.</p><h3>Detection Engineering</h3><p>Detection engineering forms the bridge between raw telemetry and actionable security incidents. After building the telemetry pipeline, the next step was developing custom Splunk detection rules capable of identifying suspicious activity across endpoint, network, and web application telemetry.</p><p>Rather than focusing on the number of detection rules, the objective was to generate meaningful alerts that could seamlessly feed into analyst investigations, automation workflows, and incident response actions.</p><p>The detections implemented throughout the lab are grouped into four primary categories.</p><h4>Windows Security Monitoring</h4><p>Windows Security Event Logs provide visibility into authentication activity across the environment.</p><p>Implemented detections include:</p><ul><li>Failed User Logins</li><li>Successful User Logins</li></ul><h4>Sysmon-Based Detections</h4><p>Sysmon significantly expands endpoint visibility by recording detailed process, network, registry, and file activity.</p><p>Implemented detections include:</p><ul><li>Credential Dumping</li><li>Suspicious PowerShell</li><li>Parent-Child Process Anomalies</li><li>Post-Exploitation Reconnaissance</li><li>Data Exfiltration</li></ul><h4>Network Detection Engineering</h4><p>Suricata provides network-level visibility into reconnaissance and intrusion attempts.</p><p>Implemented detections include:</p><ul><li>Port Scanning</li><li>Brute Force Activity</li></ul><h4>Web Application Detection Engineering</h4><p>ModSecurity with the OWASP Core Rule Set monitors HTTP traffic for common web application attacks.</p><p>Implemented detections include:</p><ul><li>SQL Injection</li><li>Cross-Site Scripting (XSS)</li><li>Command Injection</li></ul><h4>Detection Design Philosophy</h4><p>Although each detection targets a different attack technique, they were all developed with the same guiding principles:</p><ul><li>Generate actionable alerts rather than excessive noise.</li><li>Prioritize investigation value over alert volume.</li><li>Correlate multiple telemetry sources wherever possible.</li><li>Reduce false positives through meaningful detection logic.</li><li>Integrate seamlessly with automation and incident response workflows.</li></ul><p>Rather than existing as standalone Splunk searches, these detections act as the starting point of the SOC workflow. Every alert generated by Splunk is automatically processed by the automation platform, enriched with additional context, presented to analysts through Grafana dashboards, and can ultimately trigger incident response actions.</p><blockquote><strong>Note:</strong> The complete SPL queries, detection logic, MITRE ATT&amp;CK mappings, and implementation details for each detection are available in the project’s GitHub repository.</blockquote><h3>Building SOAR Automation Workflows</h3><p>Generating detections is only the beginning of the SOC lifecycle. In a busy Security Operations Center, analysts often spend more time performing repetitive administrative tasks than investigating the alerts themselves. Updating case records, enriching indicators, sending notifications, maintaining dashboards, and tracking incident status can quickly become time-consuming when performed manually.</p><p>To address this challenge, I integrated <strong>n8n</strong> as the automation platform for the lab. Rather than treating it as a simple workflow engine, n8n serves as the central orchestration layer that connects Splunk, Grafana, VirusTotal, Slack, and LimaCharlie into a single automated investigation pipeline.</p><p>The result is a SOC workflow where alerts move automatically from detection to investigation, enrichment, and response with minimal manual intervention.</p><h4>Why Automation Matters</h4><p>Modern SOCs generate hundreds or even thousands of alerts each day. Without automation, analysts can quickly become overwhelmed by repetitive tasks that add little investigative value.</p><p>The automation workflows developed for this project focus on three primary objectives:</p><ul><li><strong>Reduce analyst workload</strong> by automating repetitive tasks.</li><li><strong>Ensure consistent investigations</strong> by following standardized workflows.</li><li><strong>Accelerate incident response</strong> by enriching alerts and coordinating downstream actions automatically.</li></ul><p>Instead of manually moving information between multiple platforms, analysts can focus on understanding the incident and making informed decisions.</p><h4>n8n Workflow Architecture</h4><p>The automation platform consists of several independent workflows, each responsible for a specific stage of the SOC lifecycle. Together, these workflows transform raw Splunk alerts into structured investigations and coordinated response actions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RXzqQ1xnO9MM9SPPLOlZXw.png" /><figcaption>Figure 5. n8n workflows</figcaption></figure><h4>Alert Intake Workflow</h4><p>Acts as the primary entry point for all Splunk detections.</p><p>It receives newly generated alerts, assigns a unique SOC identifier, initializes the case lifecycle, enriches metadata, and stores the information for downstream workflows.</p><h4>Alert Dashboard API</h4><p>Provides Grafana with structured alert data, enabling the SOC dashboard to display real-time alert queues, filtering, prioritization, and analyst metrics.</p><h4>Alert Timeline API</h4><p>Maintains the investigation timeline for each case by recording analyst actions, status changes, and workflow events throughout the incident lifecycle.</p><h4>VirusTotal Enrichment Workflow</h4><p>Automatically enriches indicators of compromise such as IP addresses, domains, URLs, and file hashes with external threat intelligence.</p><p>This additional context helps analysts quickly assess the reputation of observed artifacts without leaving the investigation workflow.</p><h4>Daily Summary Bot</h4><p>Aggregates SOC activity throughout the day and generates automated summaries, providing analysts with an overview of alerts, investigations, and response actions.</p><h4>Host Isolation Workflow</h4><p>Integrates with LimaCharlie to automate endpoint containment actions once an incident has been validated.</p><p>Rather than switching between multiple platforms, analysts can initiate response actions directly from the SOC workflow.</p><h3>Building a SOC Case Lifecycle</h3><p>One of the primary goals of this project was to move beyond simple alert generation and simulate how security incidents are managed throughout their lifecycle within a Security Operations Center.</p><p>Rather than treating alerts as isolated events, every alert follows a structured workflow that reflects the progression of a real SOC investigation.</p><p>The case lifecycle includes the following states:</p><ul><li><strong>Open </strong>— A newly generated alert is received from Splunk and automatically ingested into the SOC workflow.</li><li><strong>Investigating</strong> — An analyst reviews the alert, gathers context, performs enrichment, and determines whether further action is required.</li><li><strong>Escalated</strong> — Alerts requiring additional expertise or higher-level analysis are escalated for further investigation.</li><li><strong>Closed</strong> — The investigation is completed, appropriate response actions have been taken if necessary, and the case is resolved.</li><li><strong>False Positive</strong> — Alerts determined to be benign are closed as false positives, helping improve future detection tuning and reducing analyst workload.</li></ul><p>Maintaining a structured case lifecycle enables the SOC dashboards to accurately track investigation progress, analyst activity, response times, and overall operational performance while ensuring that every alert follows a consistent investigation process.</p><h3>Automation in Action</h3><p>Rather than functioning as isolated scripts, the automation workflows operate together as a continuous pipeline.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Xm5F4jm8UuUu6Jb8P080xw.png" /><figcaption>Figure 6. Automation in Action</figcaption></figure><p>This orchestration ensures that every alert follows a consistent process from the moment it is generated until the incident is resolved.</p><h3>SOC Monitoring and Investigation Dashboards</h3><p>Collecting telemetry, engineering detections, and automating workflows are only part of the equation. Analysts still need a centralized interface to monitor alerts, prioritize investigations, and understand the overall security posture of the environment.</p><p>To support these operational requirements, I built a collection of Grafana dashboards that serve as the primary interface for the simulated Security Operations Center. Rather than functioning as simple visualization panels, these dashboards provide analysts with actionable insights throughout the entire incident lifecycle — from the moment an alert is generated until the investigation is completed.</p><h4>Why Dashboards Matter</h4><p>A modern SOC dashboard should do more than display alerts. It should help analysts answer critical questions such as:</p><ul><li>Which alerts require immediate attention?</li><li>Which systems are generating the most incidents?</li><li>How many investigations are currently in progress?</li><li>Are response time objectives being met?</li><li>What actions have been taken during the investigation?</li></ul><p>By centralizing this information into a single interface, analysts can spend less time navigating between tools and more time investigating security incidents.</p><h4>Grafana SOC Dashboard</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*swZCgELrCr0bPjvqbxpNsQ.png" /><figcaption>Figure 7. Grafana Dashboard-1</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oTWTuUG7glY8kINWvSy-Mg.png" /><figcaption>Figure 8. Grafana Dashboard-2</figcaption></figure><p>The main SOC dashboard provides a high-level overview of the current security posture while allowing analysts to quickly identify and prioritize active alerts. Information from Splunk is automatically synchronized through the n8n automation workflows, ensuring that the dashboard reflects the latest investigation status in near real time.</p><p>Key capabilities of the dashboard include:</p><h4>Alert Queue</h4><p>Displays all active alerts in a centralized queue, allowing analysts to quickly identify new incidents requiring investigation.</p><h4>Alert Prioritization</h4><p>Alerts are categorized by severity, helping analysts focus on the highest-risk events first.</p><h4>Alert Filtering</h4><p>Filtering options allow investigations to be narrowed by severity, status, host, detection type, and other contextual fields, making large alert volumes easier to manage.</p><h4>SLA Tracking</h4><p>The dashboard monitors investigation timelines and highlights alerts approaching or exceeding predefined service-level objectives, helping ensure timely incident response.</p><h4>Analyst Metrics</h4><p>Operational metrics provide visibility into SOC performance, including:</p><ul><li>Total alerts generated</li><li>Alerts by severity</li><li>Investigation status distribution</li><li>Analyst workload</li><li>Response time statistics</li><li>Case lifecycle metrics</li></ul><p>These metrics help evaluate both the effectiveness of the detection pipeline and the efficiency of analyst workflows.</p><h4>Supporting Operational Awareness</h4><p>Beyond individual investigations, the dashboards provide a real-time operational view of the SOC. Analysts can quickly assess current activity, monitor investigation progress, and identify trends without manually querying multiple systems.</p><p>By integrating data from Splunk, n8n, and the case management workflows, Grafana becomes the central workspace for monitoring day-to-day security operations.</p><h3>Building an Investigation Workflow</h3><p>Generating alerts is only the first step in incident response. Before deciding whether an alert represents a genuine security incident, analysts need sufficient context to understand what happened, which assets are involved, how severe the activity is, and what actions have already been taken.</p><p>To support this process, I built a dedicated <strong>Alert Investigation Dashboard</strong> in Grafana that serves as the primary workspace for investigating individual alerts. Rather than requiring analysts to switch between multiple platforms, the dashboard consolidates relevant information into a single investigation view.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6nWet9vWBHQQBUJwfhoK_g.gif" /><figcaption>Figure 9. Alert Investigation Dashboard used throughout the investigation lifecycle.</figcaption></figure><h4>Alert Context</h4><p>The investigation begins with a concise overview of the alert itself.</p><p>This section provides analysts with the essential information required to understand the event, including:</p><ul><li>Alert ID</li><li>Alert Title</li><li>Detection Rule</li><li>Severity</li><li>Current Status</li><li>Timestamp</li><li>Source Host</li><li>Assigned Analyst</li></ul><p>Having this information immediately available allows analysts to quickly assess the nature of the alert before beginning a deeper investigation.</p><h4>Asset Context</h4><p>Understanding the affected system is equally important as understanding the alert.</p><p>The Asset Context section provides additional information about the impacted endpoint, helping analysts determine the potential business impact of the incident.</p><p>Examples of contextual information include:</p><ul><li>Hostname</li><li>Operating System</li><li>IP Address</li><li>Logged-in User</li><li>Device Status</li><li>Asset Risk Information</li></ul><p>This additional context helps prioritize investigations involving critical systems.</p><h4>Threat Context</h4><p>Raw alerts often lack sufficient information for analysts to determine whether an event is malicious.</p><p>The Threat Context section enriches alerts with external intelligence gathered during the automation phase, allowing analysts to better understand the observed indicators.</p><p>This includes information such as:</p><ul><li>IOC Reputation</li><li>VirusTotal Results</li><li>Detection Confidence</li><li>Risk Score</li><li>Related Threat Indicators</li></ul><p>Providing this information directly within the dashboard reduces the need to manually query external threat intelligence platforms.</p><h4>Investigation Timeline</h4><p>Every action performed during the investigation is recorded in a chronological timeline.</p><p>This provides complete visibility into the progress of the investigation, including:</p><ul><li>Alert Creation</li><li>Investigation Started</li><li>IOC Enrichment Completed</li><li>Status Updates</li><li>Escalation Events</li><li>Response Actions</li><li>Case Closure</li></ul><p>Maintaining an investigation timeline improves collaboration between analysts and provides a clear audit trail for each incident.</p><h4>Analyst Notes</h4><p>Investigations often require observations that cannot be automatically generated.</p><p>The Analyst Notes section allows investigators to document findings, hypotheses, evidence collected, and decisions made throughout the investigation.</p><p>These notes become part of the permanent case record and support knowledge sharing between analysts.</p><h4>Escalation Workflow</h4><p>Not every alert can be resolved by the initial analyst.</p><p>When additional expertise is required, investigations can be escalated while preserving the complete investigation history, ensuring continuity throughout the incident lifecycle.</p><p>The dashboard reflects the current investigation status and makes it easy to identify cases that require additional attention.</p><h4>Supporting Analyst Decision-Making</h4><p>The objective of the investigation dashboard is not simply to display information — it is to help analysts make informed decisions quickly.</p><p>By combining alert details, asset information, threat intelligence, investigation history, and response actions into a single interface, the dashboard minimizes context switching and enables analysts to move efficiently from detection to containment.</p><p>Rather than navigating multiple tools to gather evidence, analysts can perform much of the investigation directly from the dashboard, allowing them to focus on understanding the incident instead of locating information.</p><h3>Threat Intelligence Enrichment</h3><h4>Integrating VirusTotal</h4><p>One of the challenges analysts face during an investigation is determining whether an observed indicator is malicious or simply unfamiliar. Searching for IP addresses, domains, URLs, or file hashes manually can interrupt the investigation workflow and increase response times.</p><p>To streamline this process, I integrated <strong>VirusTotal</strong> into the SOC automation pipeline using n8n. Whenever supported indicators are identified within an alert, they are automatically submitted for reputation lookups, and the enrichment results are attached to the investigation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eINvWcRnd0S-oAt8oAdV6Q.png" /><figcaption>Figure 10. Virustotal Enrichment API Workflow</figcaption></figure><p>The enrichment process provides analysts with additional context, including:</p><ul><li>File hash reputation</li><li>IP address reputation</li><li>Domain reputation</li><li>URL analysis</li><li>Detection ratios across multiple security vendors</li><li>Direct links to VirusTotal reports for deeper investigation</li></ul><p>By enriching alerts automatically, analysts can evaluate the trustworthiness of observed indicators without leaving the SOC investigation workflow.</p><h4>Why Context Matters More Than Raw Alerts</h4><p>A security alert only indicates that suspicious activity has been observed — it rarely provides enough information to determine whether an incident is truly malicious.</p><p>Threat intelligence enrichment transforms isolated alerts into investigations supported by additional evidence. Instead of manually pivoting between multiple external platforms, analysts receive relevant reputation data alongside the alert, enabling faster and more informed decision-making.</p><p>Within this lab, VirusTotal enrichment acts as another layer of context that complements the telemetry collected from Splunk, allowing investigations to progress more efficiently from detection to containment.</p><h3>Incident Response Automation</h3><p>Detecting and investigating security incidents are essential responsibilities of a Security Operations Center, but the ultimate objective is to contain threats before they can cause further damage. Once an analyst has gathered sufficient evidence and determined that an alert represents a genuine security incident, the SOC must transition from investigation to response.</p><p>To simulate this stage of the incident lifecycle, I integrated <strong>LimaCharlie EDR</strong> into the environment, allowing response actions to be initiated directly from the SOC workflow. Rather than manually switching between multiple security platforms, analysts can trigger endpoint containment actions as part of the same investigation process.</p><h4>From Detection to Containment</h4><p>Throughout the project, every incident follows a structured workflow from the moment suspicious activity is detected until the investigation is completed.</p><p>The response process includes the following stages:</p><ol><li><strong>Alert Generated</strong> — Splunk detects suspicious activity and creates a new security alert.</li><li><strong>Investigation Started</strong> — The alert is automatically ingested into the SOC workflow and assigned for investigation.</li><li><strong>Threat Intelligence Enrichment</strong> — Indicators of compromise are enriched using VirusTotal to provide additional context.</li><li><strong>Analyst Review</strong> — The investigation dashboard is used to examine telemetry, asset information, and threat intelligence before making a response decision.</li><li><strong>Host Isolation</strong> — If malicious activity is confirmed, the affected endpoint can be isolated through LimaCharlie to prevent further communication with the network.</li><li><strong>Containment and Recovery</strong> — Once the threat has been contained, remediation activities can begin.</li><li><strong>Case Closure</strong> — The investigation is documented, the incident is resolved, and the case is closed within the SOC workflow.</li></ol><p>This structured process ensures that every incident follows a consistent lifecycle from detection through containment and resolution.</p><h4>LimaCharlie Integration</h4><p>One of the primary response capabilities implemented in this project is automated host isolation using LimaCharlie.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6_RxzGeNP27JMfqTYBkcZA.png" /><figcaption>Figure 11. LimaCharlie Host Isolated</figcaption></figure><p>When an analyst determines that a compromised endpoint should be contained, the response workflow communicates directly with the LimaCharlie API to execute the required action.</p><p>This integration allows the SOC to move beyond passive monitoring and demonstrate an active incident response capability.</p><h4>Host Isolation</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8tghagDQzj_GNlBYFKDquw.png" /><figcaption>Figure 12. Slack Host Isolation</figcaption></figure><p>The Host Isolation workflow immediately restricts network communication for the affected endpoint, limiting an attacker’s ability to move laterally or continue interacting with compromised systems while the investigation continues.</p><h4>Host Unisolation</h4><p>Once remediation has been completed and the endpoint is considered safe, the host can be returned to normal operation through the same automated workflow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jf1QHTVphIjswmvcCjgYIQ.png" /><figcaption>Figure 13. Slack Host Unisolation</figcaption></figure><p>Automating both containment and recovery helps ensure that response actions remain consistent while reducing manual operational overhead.</p><h4>Response Workflow Automation</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pqQxCh6vFMg1hoqYMs-aSg.png" /><figcaption>Figure 14. Host Isolation Action Workflow</figcaption></figure><p>Rather than functioning as an isolated automation, the response workflow integrates with the broader SOC pipeline.</p><p>Alerts generated by Splunk progress through enrichment, investigation, analyst review, and finally response, creating a continuous end-to-end incident handling process.</p><p>By connecting Splunk, n8n, Grafana, VirusTotal, Slack, and LimaCharlie, the project demonstrates how multiple security platforms can be orchestrated into a unified response workflow.</p><h4>Measuring Response Effectiveness</h4><p>Effective incident response is not only about containing threats — it is also about understanding how efficiently incidents are handled.</p><p>Throughout the investigation lifecycle, the dashboards capture operational metrics such as:</p><ul><li>Time to Investigation</li><li>Time to Resolution</li><li>Current Case Status</li><li>Investigation Timeline</li><li>Analyst Activity</li><li>Response Actions Performed</li></ul><p>These metrics provide valuable insight into SOC performance while demonstrating how automation can improve consistency and reduce response times.</p><h3>Integrating a Phishing Attack Simulation into the SOC</h3><h4>Why Attack Simulation Matters</h4><p>Building a Security Operations Center is only part of the challenge. The true test of any SOC is its ability to detect, investigate, and respond to realistic attack scenarios.</p><p>To validate the end-to-end workflow, I integrated the phishing simulation environment that I developed in my previous projects into the SOC lab. Rather than generating synthetic alerts, this approach allowed the environment to produce realistic telemetry across the endpoint, network, and web layers while exercising every stage of the investigation and response pipeline.</p><p>For readers interested in the complete phishing infrastructure and attack simulation, I have covered those topics in detail in the following articles:</p><ul><li><a href="https://medium.com/@harikrishnan.p097/i-built-a-phishing-lab-to-finally-understand-the-cyber-kill-chain-9aad7c10e8c4"><strong>I Built a Phishing Lab to Finally understand the Cyber Kill Chain</strong></a></li><li><a href="https://medium.com/@harikrishnan.p097/inside-my-phishing-simulation-lab-red-team-meets-blue-team-05731f401bb9"><strong>Inside My Phishing Simulation Lab: Red Team Meets Blue Team</strong></a></li></ul><p>This article focuses on how those attacks are detected, investigated, and responded to within the SOC.</p><h4>Phishing Infrastructure Overview</h4><p>The attack simulation combines several components to emulate a realistic phishing campaign.</p><p>The environment consists of:</p><ul><li><strong>Gophish</strong> for phishing campaign management and email delivery.</li><li><strong>Thunderbird</strong> as the victim’s email client.</li><li><strong>Sliver C2</strong> for post-exploitation command execution and endpoint control.</li></ul><p>Together, these components simulate the progression of an attack from initial phishing email delivery through post-exploitation activity.</p><h4>Telemetry Generated by the Attack</h4><p>One of the primary goals of the phishing simulation was not simply to compromise a system, but to generate realistic telemetry that could be observed throughout the SOC.</p><p>The attack produces multiple categories of security events, including:</p><ul><li><strong>Endpoint Activity</strong> — Process creation, PowerShell execution, registry modifications, and other Sysmon events.</li><li><strong>Windows Security Events</strong> — User authentication and account activity.</li><li><strong>Network Activity</strong> — Command-and-control communication, reconnaissance, and data transfers observed by Suricata.</li><li><strong>Command-and-Control Events</strong> — Sliver session establishment and operator actions.</li><li><strong>Data Exfiltration Events</strong> — Simulated file transfers used to validate exfiltration detections.</li></ul><p>Rather than relying on isolated log sources, the SOC receives correlated telemetry from multiple systems, providing analysts with a much more complete understanding of the attack.</p><h4>Detection Coverage</h4><p>The telemetry generated by the phishing campaign exercises multiple detection rules across the environment.</p><p>The attack triggers detections covering several stages of the attack lifecycle, including:</p><ul><li>Initial Access</li><li>Suspicious PowerShell Execution</li><li>Post-Exploitation Reconnaissance</li><li>Command and Control Activity</li><li>Data Exfiltration</li><li>Network Reconnaissance</li></ul><p>This demonstrates how a single attack can generate alerts across different telemetry sources, allowing analysts to correlate activity rather than investigating isolated events.</p><h4>Analyst Investigation Workflow</h4><p>Once the detections are generated, they automatically enter the SOC workflow.</p><p>The alerts are processed by n8n, enriched with threat intelligence, synchronized with Grafana, and presented to the analyst through the investigation dashboard.</p><p>Using the investigation dashboard, analysts can:</p><ul><li>Review alert details and severity.</li><li>Examine asset context.</li><li>Analyze enriched threat intelligence.</li><li>Track investigation progress through the timeline.</li><li>Document findings using analyst notes.</li><li>Escalate the investigation when necessary.</li><li>Initiate response actions directly from the dashboard.</li></ul><p>This demonstrates how the SOC transforms raw telemetry into a structured investigation process.</p><h4>Response Actions</h4><p>After the investigation confirms malicious activity, the incident progresses into the response phase.</p><p>Depending on the findings, the analyst can:</p><ul><li>Isolate the compromised endpoint using LimaCharlie.</li><li>Notify the SOC team through Slack.</li><li>Update the investigation status.</li><li>Record analyst actions and case notes.</li><li>Close the incident once remediation has been completed.</li></ul><p>The phishing simulation therefore validates the complete SOC workflow — from initial compromise through containment and case closure — within a single integrated environment.</p><h4>Bringing the Entire SOC Together</h4><p>What began as a phishing email ultimately exercised every major component of the environment:</p><ul><li>Gophish delivered the phishing campaign.</li><li>Thunderbird simulated the victim interaction.</li><li>Sliver established post-exploitation access.</li><li>Sysmon, Windows Event Logs, Suricata generated telemetry.</li><li>Splunk detected malicious activity.</li><li>n8n automated alert processing and enrichment.</li><li>VirusTotal provided threat intelligence.</li><li>Grafana supported analyst investigations.</li><li>LimaCharlie enabled endpoint containment.</li><li>Slack delivered operational notifications.</li></ul><p>Rather than functioning as independent tools, these technologies worked together as a unified Security Operations Center, demonstrating the complete journey from attack to response.</p><h3>Lessons Learned</h3><p>Building this project fundamentally changed how I think about Security Operations Centers. Before starting the lab, I viewed security tools as individual technologies to learn independently. By the end of the project, I realized that the real challenge lies in integrating those technologies into a cohesive workflow that supports analysts throughout the incident lifecycle.</p><p>The following lessons had the greatest impact on my understanding of modern security operations.</p><h4>Detection Engineering Is Only One Part of the SOC</h4><p>When I first began learning detection engineering, my focus was almost entirely on writing SPL queries and creating detection rules.</p><p>Over time, I realized that generating alerts is only the beginning of the investigation process. Every alert requires context, triage, enrichment, analyst review, and, when necessary, response actions before it can become a resolved incident.</p><p>This project helped me understand that effective detections are those that support investigations — not simply those that generate alerts.</p><h4>Context Matters More Than Alerts</h4><p>A single alert rarely tells the complete story.</p><p>Analysts need to understand the affected asset, associated users, related telemetry, historical activity, and external threat intelligence before making response decisions.</p><p>Integrating VirusTotal enrichment, investigation timelines, and asset context demonstrated how additional information can significantly improve both investigation quality and response confidence.</p><h4>Automation Improves Consistency</h4><p>One of the biggest challenges in security operations is ensuring that every alert follows a consistent investigation process.</p><p>Automating repetitive tasks such as case creation, threat intelligence enrichment, dashboard updates, notifications, and response orchestration reduces manual effort while ensuring that important investigation steps are not overlooked.</p><p>Automation doesn’t replace analysts — it allows them to spend more time investigating and less time performing repetitive administrative work.</p><h4>Response Capabilities Make Detections Actionable</h4><p>Many home labs stop once an alert appears in the SIEM.</p><p>I wanted this project to continue beyond detection by demonstrating how validated incidents progress into containment and recovery.</p><p>Integrating LimaCharlie for endpoint isolation transformed the project from a monitoring environment into a more complete SOC simulation capable of demonstrating the full incident response lifecycle.</p><h4>Dashboards Must Support Analyst Workflows</h4><p>While building the Grafana dashboards, I discovered that effective dashboards are not simply collections of charts and statistics.</p><p>Their purpose is to help analysts answer operational questions quickly, prioritize investigations, understand incident context, and make informed decisions.</p><p>Designing dashboards around analyst workflows proved to be far more valuable than focusing solely on visual presentation.</p><h4>Building Workflows Is More Valuable Than Learning Tools Individually</h4><p>Perhaps the biggest lesson from this project was realizing that modern Security Operations Centers are built around workflows rather than individual technologies.</p><p>Splunk, Grafana, n8n, VirusTotal, LimaCharlie, Suricata, and the other tools each solve specific problems, but their true value emerges when they work together as part of an integrated security operation.</p><p>Building this environment taught me not only how each tool functions, but also how information flows between them to support detection, investigation, and response.</p><h3>GitHub Repository</h3><p>The complete project, including architecture documentation, detection engineering write-ups, workflow exports, Grafana dashboards, and supporting documentation, is available on GitHub.</p><p><strong>GitHub Repository: </strong><a href="https://github.com/Harikrishnan-P97/soc-automation-detection-engineering-lab"><strong>soc-automation-detection-engineering-lab</strong></a></p><blockquote>If you have suggestions, questions, or ideas for improving the lab, I’d be happy to hear your feedback. Likewise, if you’re building a similar SOC environment, I hope this project provides useful ideas and inspiration for your own learning journey.</blockquote><h3>Final Thoughts</h3><p>Building this SOC lab has been one of the most rewarding projects I’ve worked on. It brought together concepts that I had previously learned in isolation — detection engineering, SIEM, SOAR, threat intelligence, incident response, dashboard design, and attack simulation — and demonstrated how they function as part of a complete Security Operations Center.</p><p>Throughout the project, I realized that effective security operations are built around workflows rather than individual tools. Splunk, Grafana, n8n, VirusTotal, LimaCharlie, Suricata, ModSecurity, Gophish, Sliver, and the other technologies used in this lab each contribute a specific capability, but their real value comes from how they work together to support analysts from detection through containment.</p><p>This project also reinforced an important lesson: building a detection is only the beginning. A successful SOC must provide analysts with context, automate repetitive tasks, support efficient investigations, and enable timely response actions. Understanding how information flows between these stages proved to be far more valuable than simply learning how to configure each individual technology.</p><p>Although this lab simulates a production Security Operations Center, it also serves as a foundation for future work. There are many opportunities to expand the environment further, including additional detection content, threat intelligence integrations, case management capabilities, automated playbooks, cloud telemetry, and adversary emulation techniques.</p><p>More importantly, this project represents my journey from learning individual cybersecurity tools to understanding how modern Security Operations Centers operate as integrated systems. That perspective has fundamentally changed how I approach security engineering and incident response.</p><p>If you’re building your own SOC lab, my advice is simple: don’t stop after creating detections. Follow the alert through the entire lifecycle. Investigate it. Enrich it with context. Automate repetitive tasks. Respond to it. That’s where the most valuable learning happens.</p><h3>Disclaimer</h3><p>This project was created for educational and research purposes within a controlled lab environment. All attack simulations, phishing campaigns, command-and-control infrastructure, and incident response activities were performed exclusively against systems owned and managed within the lab.</p><p>No unauthorized testing was conducted against third-party systems, networks, or users. The techniques demonstrated are intended solely to support cybersecurity education, detection engineering, defensive research, and Security Operations Center training.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=41cd4542d78c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Built a Phishing Lab to Finally Understand the Cyber Kill Chain]]></title>
            <link>https://medium.com/@harikrishnan.p097/i-built-a-phishing-lab-to-finally-understand-the-cyber-kill-chain-9aad7c10e8c4?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/9aad7c10e8c4</guid>
            <category><![CDATA[homelab]]></category>
            <category><![CDATA[cyber-kill-chain]]></category>
            <category><![CDATA[red-team]]></category>
            <category><![CDATA[phishing]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Mon, 25 May 2026 08:59:47 GMT</pubDate>
            <atom:updated>2026-05-25T08:59:47.678Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Fi1qav-zQGxgNa1A3yZJDw.png" /></figure><h4>Mapping every stage of a phishing campaign to real telemetry, attacker actions, and SOC visibility</h4><p>When I first learned the Cyber Kill Chain, it felt easy to memorize but difficult to actually understand.</p><p>Reconnaissance. Weaponization. Delivery. Exploitation…</p><p>I knew the stages, but I couldn’t visualize how they appeared in a real attack.</p><p>Instead of reading another diagram, I decided to use my phishing simulation lab and map every phase of the attack lifecycle to real actions, telemetry, and detections.</p><p>This writeup walks through the Cyber Kill Chain using a hands-on phishing simulation environment where both red-team activity and blue-team visibility exist.</p><h4>The Lab Environment</h4><p>The environment used in this writeup is the same phishing simulation lab introduced in my previous article, where I built an end-to-end environment for simulating phishing attacks and observing defender visibility across multiple tools.</p><p>The lab contains:</p><ul><li>Gophish for phishing campaign delivery</li><li>Sliver C2 infrastructure</li><li>Windows victim machine</li><li>Sysmon telemetry collection</li><li>Splunk SIEM</li><li>Detection rules and monitoring dashboards</li></ul><p>Together, these components allow attack activity to be generated from the red-team side while simultaneously creating telemetry that can be investigated from the blue-team perspective.</p><p>Rather than rebuilding the entire lab setup again, this article focuses on something different: observing how each stage of the Cyber Kill Chain appears during a simulated attack.</p><p>Instead of treating the Cyber Kill Chain as a diagram to memorize, the goal here is to watch it unfold step by step and identify where defenders gain visibility.</p><h4>Understanding the Attack Flow</h4><p>Before mapping events to the Cyber Kill Chain, let’s walk through the phishing scenario used throughout this simulation.</p><p>The goal of this exercise was not simply to send a phishing email or establish a reverse shell. Instead, the objective was to simulate a complete attack chain that generates observable activity across multiple systems and security tools.</p><p>The scenario follows a phishing attack from initial delivery through post-exploitation actions, allowing each phase of the Cyber Kill Chain to be tied to real telemetry generated inside the lab.</p><p>The attack begins with infrastructure preparation. The attacker creates a phishing campaign in Gophish and generates an mTLS payload using Sliver. To make the delivery process appear realistic, a malicious ZIP archive is hosted through a Python web server.</p><p>The ZIP archive contains:</p><ul><li>Payroll_April_2026.pdf.lnk</li><li>payrollviewer.exe</li><li>Payroll_Readme.txt</li></ul><p>Rather than presenting an obvious executable, the files are intentionally disguised as payroll-related content to simulate a common social engineering scenario.</p><p>The attack flow used throughout this simulation is shown below:</p><ol><li>The attacker creates and hosts a phishing campaign using Gophish.</li><li>An mTLS payload is generated using Sliver.</li><li>A malicious ZIP archive is hosted through a Python web server.</li><li>The attacker sends the phishing email.</li><li>The victim opens the email in Thunderbird and clicks the download link.</li><li>The malicious ZIP archive is downloaded from the phishing landing page.</li><li>The victim extracts the ZIP archive into the user directory.</li><li>The victim opens Payroll_April_2026.pdf.lnk</li><li>The shortcut triggers execution of payrollviewer.exe</li><li>The payload executes and establishes a callback to the Sliver C2 server.</li><li>A Sliver session is created on the attacker machine.</li><li>The attacker performs reconnaissance commands such as whoami, hostname, and network enumeration activity.</li><li>Simulated data exfiltration activity is performed using Netcat.</li></ol><p>This sequence creates telemetry across email delivery systems, endpoint process execution, PowerShell activity, command-and-control communication, and post-exploitation actions. Those events become useful because each step can now be mapped directly to the Cyber Kill Chain and observed from a defender’s perspective.</p><p>With the attack workflow established, the next step is mapping those actions to the Cyber Kill Chain. Instead of treating the framework as theory, each phase below is tied directly to activity generated inside the lab.</p><h4>Stage 1 — Reconnaissance</h4><p>Reconnaissance is the phase where attackers gather information before launching an attack. This may include collecting employee details, identifying technologies in use, selecting targets, or understanding organizational structure.</p><p>Referring back to the attack flow, elements of steps 1–3 partially relate to Reconnaissance because target selection and social engineering planning occur before payload delivery.</p><p>In this simulation, the reconnaissance phase was intentionally simplified because the environment operated in a controlled lab. Instead of collecting information from external sources, the attacker already knew the target user and designed the campaign around a believable payroll-themed scenario.</p><p>The social engineering lure itself became part of the reconnaissance strategy.</p><p><strong>Attacker Activity:</strong></p><ul><li>Identify target email account</li><li>Select payroll theme</li><li>Create believable document naming</li><li>Plan delivery mechanism</li></ul><p><strong>Defender Visibility:</strong></p><p>Reconnaissance often produces little endpoint visibility.</p><p>Possible defensive opportunities may include:</p><ul><li>External scanning detection</li><li>Threat intelligence monitoring</li><li>OSINT monitoring</li><li>Email reconnaissance indicators</li></ul><p>In many cases, defenders may have limited telemetry during this phase.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9AmzR2_GBkPopuaiQt32SA.png" /><figcaption>Figure 1: User and group selection for the phishing campaign</figcaption></figure><h4><strong>Stage 2 — Weaponization</strong></h4><p>Weaponization is where attackers prepare the payload and infrastructure required for delivery.</p><p>Referring back to the attack flow, steps 1–3 primarily map to the Weaponization phase.</p><p>In this simulation, the attacker generated an mTLS payload using Sliver and prepared supporting files required for execution.</p><p>A malicious ZIP archive was then created containing:</p><ul><li>Payroll_April_2026.pdf.lnk</li><li>payrollviewer.exe</li><li>Payroll_Readme.txt</li></ul><p>The naming attempted to resemble a payroll document rather than a malicious executable.</p><p>The ZIP archive was hosted through a Python web server and linked to the phishing campaign.</p><p><strong>Attacker Activity</strong></p><ul><li>Generate Sliver mTLS payload</li><li>Create LNK execution chain</li><li>Package ZIP archive</li><li>Configure hosting infrastructure</li></ul><p><strong>Defender Visibility</strong></p><p>Visibility during weaponization is usually limited because activity happens entirely on attacker infrastructure.</p><p>Potential indicators:</p><ul><li>Malware intelligence feeds</li><li>Attachment analysis</li><li>Sandboxing</li><li>File reputation systems</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C9lD1VO8ghwjq0_jAStYzg.png" /><figcaption>Figure 2: Malicious Zip file contents</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sY024ZFMHSelHZNIJ6vXQw.png" /><figcaption>Figure 3: Attacker serving the malicious Zip file via a python server</figcaption></figure><h4>Stage 3 — Delivery</h4><p>Delivery represents how the payload reaches the victim.</p><p>Steps 4–6 align with the Delivery phase.</p><p>In this simulation, the phishing email was sent using Gophish and directed the victim toward a landing page containing the ZIP archive.</p><p>The victim opened the message using Thunderbird and clicked the embedded download link.</p><p>At this stage, malicious code had not yet executed. However, the attacker successfully moved the payload into the environment.</p><p><strong>Attacker Activity</strong></p><ul><li>Launch phishing campaign</li><li>Send phishing email</li><li>Deliver malicious archive</li></ul><p><strong>Defender Visibility</strong></p><p>This phase provides significantly more visibility opportunities:</p><ul><li>Email gateway telemetry</li><li>Attachment analysis</li><li>URL monitoring</li><li>User reporting</li><li>Email security alerts</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uhfEUKyFJdInx2GTSEaRkg.png" /><figcaption>Figure 4: Phishing email template</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*p5Mq8WI-fjLEYY6zt7SF-g.png" /><figcaption>Figure 5: Thunderbird mail inbox</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AYCztW2_z1zu615hFMI0kA.png" /><figcaption>Figure 6: Landing page for the malicious zip file download</figcaption></figure><h4>Stage 4 — Exploitation</h4><p>Exploitation begins when user interaction triggers execution.</p><p>Steps 8–9 represent Exploitation.</p><p>After downloading and extracting the ZIP archive, the victim opened: Payroll_April_2026.pdf.lnk</p><p>Instead of opening a document, the shortcut triggered execution behavior that launched: payrollviewer.exe</p><p>At this point endpoint activity became visible.</p><p><strong>Attacker Activity</strong></p><ul><li>User executes shortcut</li><li>Payload chain begins</li></ul><p><strong>Defender Visibility</strong></p><p>This stage generated valuable telemetry:</p><ul><li>Sysmon Event ID 1</li><li>Process creation activity</li><li>Parent-child relationships</li><li>Suspicious execution chain</li></ul><p>This is often where defenders begin gaining stronger visibility.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7CA1W1Ak5ADWFjkyTa5abg.png" /><figcaption>Figure 7: Victim user extracting the malicious zip file containing sliver payload</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C5ffMZwuaFjFCn45qOtH8A.png" /><figcaption>Figure 8: Zip file contents after extraction</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_Q7mLvy_KY8xc2h7SinN2w.png" /><figcaption>Figure 9: Sliver payload running as a process on the victim system</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*daq1pyMp9dkiu9Im08sPPw.png" /><figcaption>Figure 10: Gophish dashboard after the victim user executed the payload</figcaption></figure><h4>Stage 5 — Installation</h4><p>Installation typically represents the stage where attackers establish persistence or prepare long-term access mechanisms.</p><p>In this simulation, persistence techniques such as scheduled tasks, registry modifications, or service creation were intentionally excluded. Instead, installation was represented by successful payload execution and the establishment of a foothold on the victim system.</p><p>Not every attack immediately deploys persistence mechanisms, especially during early-stage compromises or controlled simulations.</p><p><strong>Defender Visibility</strong></p><p>Potential indicators may include:</p><ul><li>Registry modifications</li><li>Scheduled task creation</li><li>Service installation</li><li>Startup persistence changes</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Z3Iw1q79rnQwYVwg4QR_9A.png" /><figcaption>Figure 11: Initial foothold established on victim system</figcaption></figure><h4>Stage 6 — Command and Control</h4><p>Following execution, the payload established communication with the Sliver server.</p><p>Steps 10–11 correspond to Command and Control.</p><p>Once communication succeeded, an active session appeared inside the attacker console.</p><p>This transformed execution into an interactive compromise.</p><ul><li>Attacker Activity</li><li>Payload callback</li><li>Session creation</li><li>C2 communication</li></ul><p><strong>Defender Visibility</strong></p><p>This phase often produces:</p><ul><li>Unusual outbound connections</li><li>DNS activity</li><li>Network anomalies</li><li>Beaconing patterns</li></ul><p>Telemetry generated in the lab allowed command-and-control communication to become observable.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b1IONDtVqofvQseEXByqEA.png" /><figcaption>Figure 12: Command and control session established in sliver</figcaption></figure><h4><strong>Stage 7 — Actions on Objectives</strong></h4><p>The final phase represents attacker goals.</p><p>After obtaining access, the attacker interacted with the session and performed reconnaissance commands including:</p><ul><li>whoami</li><li>hostname</li><li>ipconfig</li></ul><p>After initial system reconnaissance, simulated data exfiltration activity was performed using Netcat to represent an attacker objective beyond access alone.</p><p>These actions represented post-exploitation objectives after initial access was achieved.</p><p><strong>Defender Visibility</strong></p><p>This phase often generates:</p><ul><li>Command execution telemetry</li><li>Process activity</li><li>Network transfers</li><li>Unusual file movement</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ahUaIUabuHccwo1NP2ILCw.png" /><figcaption>Figure 13: Execution of recon commands on the target system</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tBrtprNU03clPRiIm5wlsQ.png" /><figcaption>Figure 14: Data exfiltration activity using netcat.exe</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FvwXqLXOR1mKUiPNHkxrAg.png" /><figcaption>Figure 15: Netcat listener on attacker machine to receive data</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zrqfiVuhpTROLi92PmLtAA.png" /><figcaption>Figure 16: Successfully exfiltrated data from the victim user system</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*unoXPzb5cs02o8IAk-5i0g.png" /><figcaption>Figure 17: Splunk alerts for the data exfiltration activity</figcaption></figure><h4>Mapping Telemetry to the Cyber Kill Chain</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t_oGwqa4bKoCf2HxPRAljw.png" /><figcaption>Figure 18: Cyber kill chain mapping</figcaption></figure><h4>Closing thoughts</h4><ul><li>The Cyber Kill Chain becomes much easier to understand when you stop treating it as a framework to memorize and start treating it as something you can observe.</li><li>Before building this lab, the stages existed mostly as terms in slides and diagrams. I could recite them in order, but I struggled to connect them to actions taking place during a real attack.</li><li>Building a phishing simulation environment changed that.</li><li>Instead of reading about attack theory, I could watch every phase unfold and see the artifacts it produced: phishing emails being delivered, processes spawning on endpoints, command-and-control communication being established, and telemetry flowing into monitoring systems.</li><li>What once felt like abstract terminology became visible activity with logs, alerts, and investigation opportunities.</li><li>More importantly, the exercise highlighted where defenders gain visibility and where blind spots still exist.</li><li>Learning the Cyber Kill Chain through a lab environment transformed it from a memorization exercise into something much more practical: a way to understand attacker behavior and defender visibility at the same time.</li><li>This environment will continue evolving through additional detection logic, new phishing scenarios, SOC investigation workflows, and automated response experiments.</li><li>Because building the lab turned out to be only the beginning.</li></ul><h4>Disclaimer</h4><ul><li>This lab environment was created solely for educational and defensive research purposes. All phishing activity, payload execution, and command-and-control communication occurred within an isolated environment under controlled conditions.</li><li>The techniques demonstrated in this article are intended to help understand attacker behavior, improve detection visibility, and support blue-team learning. They should never be used against systems or environments without explicit authorization.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9aad7c10e8c4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Inside My Phishing Simulation Lab: Red Team Meets Blue Team]]></title>
            <link>https://medium.com/@harikrishnan.p097/inside-my-phishing-simulation-lab-red-team-meets-blue-team-05731f401bb9?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/05731f401bb9</guid>
            <category><![CDATA[blue-team]]></category>
            <category><![CDATA[phishing]]></category>
            <category><![CDATA[sliverc2]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <category><![CDATA[gophish]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Tue, 19 May 2026 15:43:13 GMT</pubDate>
            <atom:updated>2026-05-19T15:43:13.036Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4IT47ZHhjFS3sOUHjzmt5A.jpeg" /></figure><h4>Simulating attacks • Generating telemetry • Building detections</h4><p>Phishing continues to be one of the most common initial access vectors used in real-world cyber attacks. Despite advances in email security, endpoint protection, and user awareness programs, attackers still rely heavily on phishing because it targets the weakest link in many environments: human interaction.</p><p>A single email attachment or malicious link can become the starting point for credential theft, malware execution, command-and-control activity, and eventually lateral movement inside an environment.</p><p>While learning about phishing attacks through writeups and videos is useful, I wanted to understand the process from a different perspective:</p><h4>What actually happens after a user clicks?</h4><p>Rather than only reading about attack chains, I wanted to simulate them in a controlled and isolated environment where I could:</p><ul><li>Build realistic phishing scenarios</li><li>Observe artifacts generated on systems</li><li>Generate telemetry across multiple layers</li><li>Study attacker behavior</li><li>Build detections and monitoring around those activities</li></ul><p>The result was a phishing simulation lab that combines both red-team and blue-team perspectives.</p><p>The goal was not simply to deliver a payload. The goal was visibility.</p><p>I wanted an environment where I could simulate phishing attacks, watch activity unfold, collect logs, trigger detections, and eventually automate parts of the response process.</p><h4>Why I Built This Lab</h4><p>The lab was designed around several learning objectives:</p><p><strong>Understand attack lifecycle behavior</strong></p><p>Rather than studying isolated tools individually, I wanted to observe how an attack progresses from delivery to execution and post-exploitation activity.</p><p><strong>Generate realistic telemetry</strong></p><p>Detection engineering becomes significantly more useful when built around actual activity rather than synthetic examples.</p><p>By executing controlled phishing simulations, I could generate process creation events, network connections, child processes, and other telemetry that resembles real-world attacker behavior.</p><p><strong>Create repeatable experiments</strong></p><p>One-off demonstrations are useful, but repeatable environments are far more valuable.</p><p>This setup allows me to rerun campaigns, modify payload behavior, test detection logic, and compare outcomes.</p><p><strong>Support future SOC projects</strong></p><p>This lab also acts as a foundation for future projects involving:</p><ul><li>Detection engineering</li><li>Incident response workflows</li><li>SOC automation</li><li>Alert enrichment</li><li>Dashboard creation</li><li>Investigation pipelines</li></ul><h4>Architectural Overview</h4><p>At a high level, the environment consists of:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CSr7eljd7ERiW-HqfRjLrQ.png" /><figcaption>High level overview of the environment</figcaption></figure><p>The phishing execution flow looks like this:</p><figure><img alt="Phishing simulation architecture diagaram" src="https://cdn-images-1.medium.com/max/1024/1*GwtSn61fim57Dou8Jpf6sA.png" /><figcaption>Execution flow of the phishing attack</figcaption></figure><h4>Lab Components</h4><p>The environment is composed of three primary sections: the attacker system, victim system, and monitoring infrastructure. Each component serves a specific purpose in simulating the phishing lifecycle while maintaining visibility into the resulting activity.</p><p><strong>Attacker System</strong></p><p>The attacker VM is responsible for simulating phishing delivery and post-exploitation behavior.</p><p>The goal was not to recreate a full offensive infrastructure but to generate realistic activity that can later be observed and detected.</p><p>Tools used:</p><ul><li><strong>Gophish: </strong>phishing campaign creation and email delivery</li><li><strong>Sliver C2: </strong>command-and-control framework used for callback simulation</li></ul><p>This system acts as the starting point of the attack chain and drives the initial access simulation.</p><p><strong>Victim System</strong></p><p>The victim machine simulates a user workstation that receives and interacts with phishing content.</p><p>To create realistic telemetry, multiple monitoring components were installed on the endpoint.</p><p>Components used:</p><ul><li><strong>Thunderbird: </strong>email client used for phishing delivery and interaction</li><li><strong>Sysmon: </strong>process, network, and endpoint activity monitoring</li><li><strong>Splunk Universal Forwarder: </strong>forwards endpoint logs to the SIEM platform</li></ul><p>This endpoint serves as the primary telemetry generation source throughout the simulation.</p><p><strong>Monitoring System</strong></p><p>The monitoring environment provides visibility into activity generated across the attack chain.</p><p>Collected telemetry is ingested into Splunk where detections and dashboards provide operational visibility.</p><p>Key capabilities include:</p><ul><li>Log collection and normalization</li><li>Detection engineering</li><li>Alert generation</li><li>Dashboard visualizations</li><li>Investigation workflows</li></ul><p>The monitoring stack transforms raw events into actionable data.</p><h4>Telemetry and Detection Opportunities</h4><p>One of the primary goals of the lab was visibility.</p><p>Rather than focusing only on payload execution, I wanted to observe the artifacts generated at different stages of the phishing lifecycle.</p><p>This setup allows monitoring of activities such as:</p><ul><li>Process creation events</li><li>Network connections</li><li>Suspicious parent-child process relationships</li><li>Attachment execution behavior</li><li>Encoded PowerShell activity</li><li>Persistence mechanisms</li><li>Command-and-control callbacks</li></ul><p>These artifacts become useful for creating detections, building alerts, and improving investigation workflows.</p><h4>Quick Demo and Screenshots</h4><p>The following sections demonstrate the phishing workflow from delivery through monitoring and alert generation.</p><p><strong>Gophish Campaign</strong></p><p>The phishing campaign was created and launched using Gophish.</p><p>The campaign included a phishing email, a payload delivery mechanism, and tracking for recipient interaction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VfosMK-rV-srYdnPtq-TFA.png" /><figcaption>Gophish phishing campaign dashboard</figcaption></figure><p><strong>Email Delivery</strong></p><p>The phishing email was delivered to the victim system using Thunderbird.</p><p>This stage simulates how users receive and interact with potentially malicious content in a real environment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*p5Mq8WI-fjLEYY6zt7SF-g.png" /><figcaption>Phishing email with link on thunderbird mail client</figcaption></figure><p><strong>Payload Interaction</strong></p><p>The simulated victim downloaded the attachment, extracted the contents, and executed the payload.</p><p>The archive contained social-engineering themed files designed to mimic realistic user behavior.</p><p>Observed activities included:</p><ul><li>Downloading the archive</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AYCztW2_z1zu615hFMI0kA.png" /><figcaption>Landing page for the malicious zip file download</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sY024ZFMHSelHZNIJ6vXQw.png" /><figcaption>Attacker serving the malicious zip file</figcaption></figure><ul><li>Extracting the files</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7CA1W1Ak5ADWFjkyTa5abg.png" /><figcaption>Zip file extraction on to the user directory</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C5ffMZwuaFjFCn45qOtH8A.png" /><figcaption>Extracted files</figcaption></figure><ul><li>Executing the shortcut</li></ul><p>When the user opens Payroll_April_2026.pdf.lnk, the shortcut launches payrollviewer.exe, which establishes a callback connection to the Sliver C2 infrastructure.</p><p>This activity subsequently generates endpoint telemetry that can be observed through Sysmon and Splunk.</p><ul><li>Payload process execution</li></ul><p>Following execution of the shortcut, the payrollviewer.exe process becomes active on the victim system and begins communication with the Sliver C2 infrastructure.</p><p>From a monitoring perspective, this stage becomes particularly important because process execution creates artifacts that can later be observed through endpoint telemetry.</p><p>Examples include:</p><ul><li>Process creation activity</li><li>Parent-child process relationships</li><li>Network connections</li><li>Command execution behavior</li></ul><p>The process can be observed running on the endpoint through Task Manager and subsequently generates telemetry collected by Sysmon.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_Q7mLvy_KY8xc2h7SinN2w.png" /><figcaption>Malicious executable running as a process on the victim machine</figcaption></figure><ul><li>Callback activity generation</li></ul><p>After execution, the payload successfully established a callback connection to the Sliver C2 server, resulting in an active session on the attacker system.</p><p>Once the session was opened, commands could be executed on the victim machine, simulating post-exploitation activity and generating additional telemetry for monitoring and detection.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b1IONDtVqofvQseEXByqEA.png" /><figcaption>Command and control session opened on sliver after payload execution</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ahUaIUabuHccwo1NP2ILCw.png" /><figcaption>Command execution on victim machine via sliver c2</figcaption></figure><p>Gophish also recorded user interaction events.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PyEXB7Q50JkVrCsKbESsaA.png" /><figcaption>Gophish dashboard after the victim user has downloaded the malicious zip file</figcaption></figure><p><strong>Splunk Alerts Triggered</strong></p><p>Telemetry forwarded through Splunk Universal Forwarder was ingested into Splunk and evaluated against detection logic.</p><p>Several alerts were generated from the observed activity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gws-JcUL6UzOkAHZLYnTNQ.png" /><figcaption>Alerts triggered on splunk</figcaption></figure><p><strong>Alert Queue and Dashboard Visualizations</strong></p><p>Beyond alert generation, telemetry was integrated into SOC dashboards and investigation workflows.</p><p>Rather than treating detections as isolated events, alerts were forwarded into a dashboard environment designed to simulate a SOC analyst workflow. This provides visibility into alert prioritization, investigation context, and activity timelines.</p><p>The dashboards allow analysts to:</p><ul><li>View active alerts in a centralized queue</li><li>Track investigation activity through timelines</li><li>Review alert context and associated artifacts</li><li>Monitor statistics and workflow metrics</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MMMojgpfDQlcNpXr1o9mHg.png" /><figcaption>SOC alert queue on grafana</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Uzqtq73EtMVqR-SPnLsZSQ.png" /><figcaption>Suspicious file creation in user directory alert overview</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TrHWd2rG3N4lKV0N6X3RQg.png" /><figcaption>Suspicious archive extraction activity alert overview</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lvTEbpKQv7LTgfaNPCUMDw.png" /><figcaption>Suspicious process execution from user directory alert overview</figcaption></figure><h4>Closing Thoughts</h4><p>This lab is not intended to be a one-time exercise.</p><p>Instead, it serves as the foundation for future experiments involving phishing detection, incident response workflows, cyber kill chain analysis, and SOC automation projects.</p><p>Building the environment provided significantly more value than simply reading about phishing attacks because it created a repeatable process for generating telemetry, validating detections, and experimenting with monitoring strategies.</p><p>Future work will expand this environment with additional detection logic, investigation workflows, alert enrichment, and automation use cases.</p><p>If you made it this far, thanks for reading. Additional writeups expanding on detections, monitoring workflows, and SOC automation built around this environment will follow.</p><h4>Disclaimer</h4><p>This project was built in an isolated lab environment strictly for educational and defensive research purposes.</p><p>No systems outside the controlled environment were targeted or impacted.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=05731f401bb9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Writer Behind the Packets: A Quick Hello]]></title>
            <link>https://medium.com/@harikrishnan.p097/the-writer-behind-the-packets-a-quick-hello-006c3b6ed0ff?source=rss-a542fa5a502b------2</link>
            <guid isPermaLink="false">https://medium.com/p/006c3b6ed0ff</guid>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[security-operation-center]]></category>
            <category><![CDATA[learning]]></category>
            <category><![CDATA[blue-team]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Harikrishnan P]]></dc:creator>
            <pubDate>Fri, 15 May 2026 04:49:36 GMT</pubDate>
            <atom:updated>2026-05-15T04:49:36.171Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*o0AUuCSGKBuDuQKB" /><figcaption>Photo by <a href="https://unsplash.com/@hieu101193?utm_source=medium&amp;utm_medium=referral">Đào Hiếu</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h4><em>A security researcher, bug hunter, and open source builder walks into a blog…</em></h4><p>I have spent years learning how attackers think, how defenders respond, and somewhere along the way I picked up a pen. This blog is what happens when those two worlds collide.</p><h4>Who I am</h4><p>Hi, I am Harikrishnan P, a red/blue teamer, security researcher, and cybersecurity writer based in India. With a background in computer science, I have spent the past few years building detection logic, diving deep into web application vulnerabilities, and recently took the plunge into bug bounty hunting.</p><p>I sit somewhere between offense and defense and I wouldn’t have it any other way. I am currently looking for SOC Analyst L1 roles, and this blog is part of that journey made public.</p><h4>What I do</h4><p>I spend my time exploring both sides of cybersecurity, from blue-team workflows and detection engineering to offensive security labs, CTFs, and practical experiments. Along the way I build projects, break things, understand how they work, and document what I learn.</p><h4>Why I write</h4><p>Writing is how I truly learn. If I cannot explain it, I do not really know it. But beyond that, this blog is a debt I am paying back. The community that shaped me did so through free resources, open write-ups, and tools built by people who asked for nothing in return. I want to be that for someone else. One person finding their footing faster because of something I wrote here that is the whole point.</p><h4>What you will find here</h4><ul><li>Lab setups and projects</li><li>Incident response breakdowns</li><li>Detection engineering deep dives</li><li>SOAR automation and playbooks</li><li>Certification experiences and honest reviews</li><li>CTF writeups and walkthroughs</li><li>Interview experiences and lessons</li><li>And everything in between as I continue learning and building.</li></ul><h4>Who is this for</h4><ul><li><strong>For fellow learners and beginners</strong>: if you are figuring out cybersecurity, switching careers, or just curious, this is your space. I am not writing from the top of the mountain. I am writing from the trail.</li><li><strong>For recruiters: </strong>what you will see here is not a polished portfolio. It is a record of how I think. My write-ups, my lab notes, my honest reflections on interviews. All of it shows how I approach problems, how I communicate, and how I grow.</li></ul><p>If any of this resonates with you, hit follow and stick around. There is a lot more coming. You can also find me here:</p><ul><li><a href="https://www.linkedin.com/in/harikrishnanp097/">LinkedIn</a></li><li><a href="https://github.com/Harikrishnan-P97">GitHub</a></li></ul><p><em>What got you into cybersecurity? Drop it in the comments. I would love to know your origin story.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=006c3b6ed0ff" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>