# Leo Nelson Personal blog of Leo Nelson talking about technology, education, customer relationship management, customer service and Africa. - Site: https://leonelson.com/ ## On Excellence - Link: https://leonelson.com/2026/09/on-excellence/ - Published: 2026-09-28T12:00:16-04:00 - Author: Leonard Nelson > The pursuit of excellence does not need justification.[](https://x.com/mitchellh) > [Mitchell Hashimoto](https://x.com/mitchellh/status/2074225453217505494?lang=en) ## Detecting and Mitigating Active Directory Compromises - Link: https://leonelson.com/2026/09/detecting-and-mitigating-active-directory-compromises-2/ - Published: 2026-09-20T09:26:51-04:00 - Author: Leonard Nelson The [Detecting and Mitigating Active Directory Compromises guidance](https://www.cyber.gov.au/sites/default/files/2026-09/Detecting%20and%20mitigating%20Active%20Directory%20compromises%20%28September%202026%29.pdf) was authored by the Australian Signals Directorate (ASD), the Cybersecurity and Infrastructure Security Agency (CISA), the National Security Agency (NSA), the Canadian Centre for Cyber Security (CCCS), the New Zealand National Cyber Security Centre( NCSC-NZ), and the United Kingdom’s National Cyber Security Centre (NCSC-UK). It aims to inform organisations about 17 common techniques used to target Active Directory as observed by the authoring agencies. This guidance provides an overview of each technique and how it can be leveraged by malicious actors, as well as recommended strategies to mitigate these techniques. By implementing the recommendations in this guidance, organisations can significantly improve their Active Directory security, and therefore their overall network security, to prevent intrusions by malicious actors. Download: [Detecting and Mitigating Active Directory Compromises](https://leonelson.com/wp-content/uploads/2026/09/Detecting-and-mitigating-Active-Directory-compromises-September-2026.pdf) Other Referece Articles: * [Detecting and Mitigating Active Directory Compromises, Published in 2024](https://leonelson.com/2024/10/detecting-and-mitigating-active-directory-compromises/) * [AD and Azure AD Auditing](https://leonelson.com/2021/02/ad-auditing/) ## AI Assistant Benchmark - Link: https://leonelson.com/2026/09/ai-assistant-benchmark/ - Published: 2026-09-15T07:13:39-04:00 - Author: Leonard Nelson At the time of this writing, AI Assistant Benchmark had 71 assistants assessed across 15 dimensions. Notably absent from this list were [OpenClaw](https://openclaw.ai/) and [Microsoft Scout](https://www.microsoft.com/en-us/microsoft-365/blog/2026/06/02/introducing-microsoft-scout-your-always-on-personal-agent/). ![](https://leonelson.com/wp-content/uploads/2026/09/image-1024x508.png) ## On Integrity - Link: https://leonelson.com/2026/08/on-integrity/ - Published: 2026-08-12T22:00:00-04:00 - Author: Leonard Nelson > Never put yourself in a position, no matter what the inducement – whether a big > salary or other financial reward, or the promise of position or influence – where > you cannot act the part of a man. Let no consideration tie your tongue or purchase > your opinion. Regard your independence as your inalienable right, with which you > will never part for any consideration. > Orison Swett Marden ## When Scammers Borrow Our Trust - Link: https://leonelson.com/2026/07/when-scammers-borrow-our-trust/ - Published: 2026-07-15T15:55:00-04:00 - Author: Leonard Nelson A text arrives about an unpaid toll. A caller says you missed jury duty. An email promises a look at your organization’s salary information. One appeals to responsibility, another to fear, and the third to curiosity. Each gives you a reason to act before questioning the source. In a high-trust society, people generally expect others to act in good faith and institutions to operate fairly. That expectation makes everyday life easier. We cooperate, do business, and follow public processes without treating every interaction as a potential threat. It also gives scammers something valuable to imitate. On the other hand, the comparison with a “low-trust society” needs care. Lower trust does not mean lower integrity. People may have learned through experience to be cautious about strangers or institutions, placing greater confidence in family, friends, and established relationships. Both patterns can exist within the same community. For a scammer, the question is: Whose credibility will persuade this person to act? Where institutional trust is strong, an official-looking notice may carry weight. A toll message presents a small balance with a looming penalty. A jury-duty caller claims to represent the court and threatens arrest. A supposed speeding violation directs the recipient to a payment page. The message exploits our willingness to meet an obligation, amplified by the fear of consequences. Where trust is more personal, a scammer may impersonate a relative needing urgent help, take over a friend’s account, or introduce a fraudulent opportunity through a familiar community group. The apparent relationship supplies the credibility. These patterns help explain how trust can be exploited but they do not establish that one kind of society experiences more fraud. Most importantly, skepticism toward strangers does not make anyone immune to an appealing message. Sometimes, the strongest invitation is simply the chance to see something we would not ordinarily see. An email claiming to identify coworkers with COVID may tap into concern for our health. A document labeled “Salary Adjustments” may appeal to questions about fairness. A link to supposedly confidential performance evaluations may provoke anxiety about what others have said about us—or curiosity about our colleagues. A familiar sender or company logo can make that unexpected access seem plausible. Curiosity supplies the remaining push. Before we ask whether we should have the information, we have already clicked. The useful pause is to question both the source and the request: “Why am I receiving this, and why am I being asked to act this way?” An unexpected offer of confidential information deserves scrutiny. So does a demand to pay immediately, stay on the phone, keep the conversation secret, or bypass a normal process. Those details can tell us more than polished wording or an official- looking logo. Verification works best when we take control of the next step. Open your usual toll account yourself. Contact the court through independently obtained contact information. Call a relative using the number already saved in your phone. Check an unexpected workplace document with its supposed sender through a known channel. Make that habit welcome within your family and workplace: “If an unusual request appears to come from me, please check. I won’t be offended.” And if curiosity or pressure wins out, report what happened promptly. Embarrassment should never prevent someone from asking for help. Trust and curiosity both enrich our lives. A moment of independent verification helps us exercise them with care. [Written with AI assistance, with ideas, editorial review, and final approval by me]. ## Zero Trust Framework for AI Agent - Link: https://leonelson.com/2026/06/zero-trust-framework-for-ai-agent/ - Published: 2026-06-02T11:11:00-04:00 - Author: Leonard Nelson Anthropic [published](https://claude.com/blog/zero-trust-for-ai-agents) their [Zero Trust Framework for AI Agents](https://cdn.prod.website-files.com/6889473510b50328dbb70ae6/6a1611a04085d7cd3dadc924_Claude-eBook-Zero-Trust-for-AI-Agents-05182026.pdf) which defines the controls autonomous systems require. The framework provides an important reference model that turns Zero Trust principles into operational controls based on the key principles of Zero Trust: 1. Never trust and always verify: Every access request undergoes authentication and authorization regardless of origin. A request from inside the corporate network receives the same scrutiny as one from an external IP address. 2. Assume breach: Design systems while expecting that compromise will occur. Rather than focusing on preventing intrusion, limit the damage an attacker can cause. Segment by identity, implement fine-grained access controls, and ensure that compromising one system doesn’t grant access to others. 3. Least privilege: Grant only the minimum access necessary for a specific task. A database administrator doesn’t need access to the email server. By constraining what each identity can access, organizations contain the blastradius of any single compromise. ## DNS Panopticon – Visualize and Inspect DNS Zone Data - Link: https://leonelson.com/2026/03/dns-panopticon-visualize-and-inspect-dns-zone-data/ - Published: 2026-03-15T23:00:12-04:00 - Author: Leonard Nelson DNS Panopticon is a simple tool designed to help visualize and inspect DNS zone data. It takes a DNS zone export (such as a .db file from BlueCat or other DNS platforms) and renders the records into a structured grid so they can be quickly reviewed, filtered, and sanity-checked. The goal is straightforward: make DNS data easier to see, analyze, and troubleshoot. You can try it here: [https://dnspanopticon.com/](https://dnspanopticon.com/) **Why DNS Panopticon?** Anyone who works with DNS infrequently knows that it is both foundational and can at times be frustrating. DNS quietly underpins almost every system on the internet, translating human-readable domain names into IP addresses so applications can communicate. But DNS data itself is often buried inside raw zone files, configuration exports, or text formats that are hard to review at scale. When troubleshooting, validating migrations, or reviewing large zones, engineers frequently end up scrolling through long files or writing quick scripts just to answer simple questions. DNS Panopticon was built to make that process easier. Instead of manually parsing zone files, the tool loads the export and presents the records in a structured interface where they can be quickly inspected. **What It Does?** At a high level, DNS Panopticon focuses on three things: * Parse data locally on your computer and maintain privacy of your DNS export * Zone file visualization – Load DNS .db exports and view records in a structured grid. * Fast inspection – Quickly scan records, identify anomalies, and validate configurations. * Operational checks – Perform quick sanity checks when reviewing DNS data. The name Panopticon is intentional: the idea is to provide complete visibility into DNS records, a single place where everything in a zone can be observed at once. **Use Cases** * Some situations where DNS Panopticon can be useful: * Reviewing DNS exports before migrations * Quickly auditing records in large zones * Identifying unusual or inconsistent entries * Performing quick validation checks on DNS data * Troubleshooting name resolution issues If you’ve ever opened a large zone file and thought “there has to be an easier way to look at this,” that’s exactly the problem this tool aims to solve. I’d [appreciate your feedback](https://leonelson.com/contact/) and hope this helps you in some way. ## A Quiet Style Find: Fowler Wool Multi Stripe Tie - Link: https://leonelson.com/2026/03/a-quiet-style-find-fowler-wool-multi-stripe-tie/ - Published: 2026-03-08T00:18:22-05:00 - Author: Leonard Nelson ![](https://leonelson.com/wp-content/uploads/2026/03/image.png) Some pieces stand out without trying too hard. The [Fowler Wool Multi Stripe Tie from Mountain & Sackett](https://mountainandsackett.com/collections/new-arrivals/products/fowler-wool-multi-stripe-tie) is a strong example: handmade in New York City, cut from 100% Italian wool, and built around earth-toned multi-stripes that feel classic, textured, and easy to wear in cooler seasons. It fits the kind of understated menswear item worth bookmarking when thinking about where style, craftsmanship, and everyday wearability meet. ## Using Resend for Email Delivery on Your Application - Link: https://leonelson.com/2026/02/using-resend-for-email-delivery-on-your-application/ - Published: 2026-02-21T23:01:43-05:00 - Author: Leonard Nelson [Resend](https://resend.com/) is an email API built for application teams who want to ship reliable transactional (and marketing) email without turning their codebase into a mail-ops project. Why it’s a solid fit for modern apps * Drop-in delivery options: send via a REST API or SMTP relay, plus support for inbound email when you need it * Developer-friendly integrations: official SDKs across common stacks (Node.js, Python, PHP/Laravel, Ruby, Go, Java, Rust, .NET) * Operational visibility: detailed request/response logs and email event tracking so you can debug a wide variety of issues, especially the guaranteed roadblock with “why didn’t this email arrive?” * Deliverability fundamentals baked in: you verify a domain you own via SPF + DKIM, with guidance to use a subdomain to isolate reputation; you can optionally add DMARC for extra trust * App-ready features: scheduled sends, batch sending, open/link tracking, templates, and webhooks for event-driven workflows (e.g., log opens/clicks, trigger follow- ups) * Security posture you can point to: pricing lists SOC 2 Type II, GDPR, API key permissions, and signed webhook endpoints Cost model that’s easy to pilot: a free tier includes 3,000 emails/month with a 100/day cap and 1 domain, then scales (e.g., $20/month for 50,000 emails on Pro). ## Engineering Excellence - Link: https://leonelson.com/2026/02/engineering-excellence/ - Published: 2026-02-01T09:53:54-05:00 - Author: Leonard Nelson > A designer knows he has achieved perfection not when there is nothing left to > add, but when there is nothing left to take away. > – Antoine de Saint-Exupéry