[{"content":"","date":null,"permalink":"https://devlogiclabs.com/blog/","section":"Blog","summary":"","title":"Blog"},{"content":"","date":null,"permalink":"https://devlogiclabs.com/","section":"DevLogicLabs","summary":"","title":"DevLogicLabs"},{"content":"A word everyone uses #\u0026ldquo;AI agent\u0026rdquo; is everywhere now. But many people use this word in different ways. Before I build my agents for CVE Friday Brief, I want to be clear about what an agent really is.\nHere is my simple definition.\nFirst, a plain LLM call #An LLM (Large Language Model) is the engine behind tools like ChatGPT. On its own, it does one thing: you send text in, and it sends text back.\nYou ask a question. It gives an answer. Then it forgets everything. This is one call, one answer. Nothing more.\nThis is useful, but it is not an agent.\nWhat turns an LLM into an agent #An agent is an LLM that can decide its own steps to reach a goal. This is the key idea. With a plain LLM call, you control every step. With an agent, the model chooses what to do next, based on what it finds along the way.\nTo do this, an agent has three parts:\nA goal. Not one question, but a task to finish. For example: \u0026ldquo;Find this week\u0026rsquo;s most important CVEs.\u0026rdquo; Tools. The agent can do actions, not just talk. It can call an API, read a database, or run a search. (Tools are optional, but most useful agents have them.) Freedom to choose. The agent works step by step. It reasons, picks an action, looks at the result, and then decides the next step on its own. It repeats until the goal is done. So the simple idea is:\nAn agent is an LLM that picks its own path to a goal, using tools when it needs them.\nAgent or workflow? #There is one more word you will hear a lot: workflow. It is easy to mix it up with an agent, so here is the simple difference.\nIn a workflow, you decide the steps. The system always runs the same path. It is predictable. In an agent, the model decides the steps at runtime. The path can change every time. It is flexible, but less predictable. Both are useful. In fact, most real systems mix the two: a fixed workflow as the \u0026ldquo;spine\u0026rdquo;, with agents handling the hard, flexible parts in the middle. I will come back to this choice in a later post.\nA small example #Imagine I ask a plain LLM: \u0026ldquo;Is Log4Shell (CVE-2021-44228) still a real threat?\u0026rdquo;\nA plain LLM can only guess from old training data. It cannot check anything new.\nAn agent works in a different way:\nIt reads the goal: check the current status of this CVE. It calls a tool to fetch live data from NIST NVD. It calls another tool to check the CISA KEV list (known exploited vulnerabilities). Log4Shell is there - attackers still use it, years after it was found. It looks at the results and writes a clear answer, based on real data. Same question. Very different quality.\nWhy this matters for my project #CVE Friday Brief is not one big question. It is a real task with many steps: collect data, choose what matters, add context, write a detection rule.\nNo single LLM call can do all of that well. But a small team of agents, each with its own goal and tools, can.\nIn the next post, I will show the full pipeline: how raw CVE data becomes a finished Friday brief.\nThanks for reading. The next post is coming soon - subscribe via RSS if you want to follow the build.\n","date":"July 22, 2026","permalink":"https://devlogiclabs.com/blog/2026-07-22/what-is-an-ai-agent/","section":"Blog","summary":"An agent is not just one LLM call. It is a model that picks its own steps and uses tools to get there. Here is the plain-English difference between a workflow and an agent, with a Log4Shell example.","title":"What Is an AI Agent? A Simple Explanation"},{"content":"Status: In development · Building in public\nCVE Friday Brief is an automated vulnerability intelligence service. It works every day in the background: it collects new CVE data from public sources, filters out the noise, and adds context. Every Friday, it generates one short report with the most important vulnerabilities of the week - what they are, why they matter, and how to detect them.\nThe interesting part is how it works: a small team of AI agents does the job. One agent selects the CVEs that matter, one enriches them with context, one writes detection rules (Sigma / KQL), and one edits everything into a clean, readable brief.\nThe goal of this project is simple: build something useful, and learn multi-agent GenAI systems by doing - with every step documented on this blog.\nData sources # NIST NVD - the main CVE database CISA KEV - known exploited vulnerabilities EPSS - exploit probability scores GitHub Security Advisories - ecosystem-level advisories Tech stack # Go - core service and data pipeline Google Agent Development Kit (ADK) - the multi-agent layer PostgreSQL - storage for CVE data and generated briefs D2 - architecture diagrams as code Hugo - this blog ","date":null,"permalink":"https://devlogiclabs.com/projects/cve-friday-brief/","section":"Projects","summary":"A multi-agent GenAI system in Go that turns the weekly flood of CVEs into one short, useful Friday report.","title":"CVE Friday Brief"},{"content":"","date":null,"permalink":"https://devlogiclabs.com/projects/","section":"Projects","summary":"","title":"Projects"},{"content":"Too many CVEs, too little time #Last week, more than 800 new CVEs (security vulnerabilities) were published. Did anyone read them all? Of course not.\nMost of them don\u0026rsquo;t matter for you. But a few of them matter a lot. The hard part is finding those few - every single week.\nI am Bartlomiej, a DevOps engineer from Poland. I work with vulnerabilities and security tools every day, and I want to use this knowledge to learn something new: GenAI agents. So I am building a system that solves this problem for me.\nThis is my new project: CVE Friday Brief.\nWhat Is CVE Friday Brief? #The service works every day: it collects new CVE data from public sources and processes it in the background.\nThen, every Friday, it generates one report - a short brief (for example, a PDF) with the most important CVEs of the week. For each one, the brief explains:\nWhat is the vulnerability? Why is it important right now? A detection rule (Sigma or KQL), when detection is possible, so a security team can find it in their logs. No long reports. Just the key facts, fast.\nOne entry in the brief could look like this (using the famous Log4Shell as an example):\nCVE-2021-44228 - RCE in Apache Log4j, \u0026ldquo;Log4Shell\u0026rdquo; (CVSS 10.0)\nStatus: In the CISA KEV list. Why now: Public exploits available, still used by attackers today. Detection: Sigma rule - look for ${jndi: strings in web request logs. How It Works #I will not write this by hand every week. Instead, I will build a small team of AI agents:\nOne agent selects the most important CVEs from public sources like NIST NVD, CISA KEV, and EPSS. One agent enriches each CVE with extra context, like public advisories. One agent writes a detection rule for it. One agent edits everything into one clean, short brief. I will build this with Go, Google\u0026rsquo;s Agent Development Kit (ADK), and PostgreSQL.\nMy Goals #I have two goals with this project:\nBuild something useful, not just a demo. Learn GenAI multi-agent systems by doing, not just by reading about them. This is a personal, open project. It is not connected to any company or product I work on - just me, building in public.\nWhat\u0026rsquo;s Next #In the next posts, I will explain:\nWhat an AI agent really is, in simple words My full pipeline, from raw data to a finished brief When I use an agent, and when plain code is enough Thanks for reading. If you want to follow this project, add the blog to your RSS reader or find me on LinkedIn - every step of the build will be documented here.\n","date":"July 9, 2026","permalink":"https://devlogiclabs.com/blog/2026-07-09/why-im-building-cve-friday-brief/","section":"Blog","summary":"Over 800 new CVEs land every week and almost none of them matter to you. I\u0026rsquo;m building a multi-agent GenAI system in Go that finds the few that do, and documenting every step here.","title":"Why I'm Building CVE Friday Brief"},{"content":"Hi, my name is Bartłomiej Nogieć. I work as an Expert DevOps Engineer at ING Hubs Poland. I have been in IT for many years. I have designed system architecture, built automation and CI/CD processes, and delivered production software for demanding clients.\nAt work and in my free time, I move between a few areas. They are different, but they often meet:\nSoftware engineering - good practices, architecture, code quality and processes. They make the road from an idea to a working product much shorter. Cybersecurity - I think about security from the start of a project, not as something you add at the end. IoT - I find it fascinating how software meets the physical world, and how to design systems that work well when they run in many places at once. Generative AI - the world of generative models changes very fast. I follow it and look for practical uses that bring real value. Aviation - my long-time passion. It teaches me a lot about reliability, procedures and safe systems. On this blog I want to share my thoughts, notes from practice, and things I learn in these areas. I hope you find something useful here - whether you write software every day, or you are just starting out.\nContact #If you want to talk - about a project, about working together, about an interesting idea, or just to share thoughts on the topics above - please get in touch. The fastest way to reach me is LinkedIn. I will be happy to reply.\n","date":null,"permalink":"https://devlogiclabs.com/about/","section":"DevLogicLabs","summary":"","title":"About me"},{"content":"Last updated: 3 May 2026\nData controller #The controller of personal data is Bartłomiej Nogieć, who runs the DevLogicLabs blog at devlogiclabs.com. Contact: LinkedIn or GitHub.\nWhat data we collect #The DevLogicLabs site does not require registration or any personal data. Data is collected automatically for analytics purposes only, and only after you give your consent.\nGoogle Analytics 4 (after consent) #After you click \u0026ldquo;Accept\u0026rdquo; in the cookie banner, the site loads the Google Analytics 4 script, which collects:\nAn anonymous client identifier (the _ga cookie) IP address (anonymised before storage - the anonymize_ip function) Device and browser information (operating system, screen resolution, language) Traffic data (pages visited, time spent on the site, referral source) Geographic location at city or region level The data is processed by Google Ireland Limited in accordance with Google\u0026rsquo;s privacy policy. Retention period: 2 months (the minimum allowed in GA4).\nWhat we do with this data #I use the Google Analytics data solely to:\nUnderstand which content is valuable to readers Improve the blog\u0026rsquo;s structure and navigation Detect technical errors (for example, broken links) The data is not sold, shared with third parties, or used for advertising profiling.\nCookies #Essential (always active) # cc-consent - stores your decision about consent to analytics. Without this cookie the banner would appear on every visit. A localStorage entry, stored indefinitely. Analytics (only after consent) # _ga, _ga_* - Google Analytics 4. Anonymous session and client identifier. Validity: 2 years. Set only after you click \u0026ldquo;Accept\u0026rdquo; in the banner. Your rights #Under the GDPR, you have the right to:\nAccess your data Rectification of inaccurate data Erasure of your data (the \u0026ldquo;right to be forgotten\u0026rdquo;) Restriction of processing Object to processing Withdraw consent at any time To withdraw your consent to Google Analytics, you can:\nClear your browser data (cookies + localStorage) for the devlogiclabs.com domain - the banner will appear again Contact me via LinkedIn or GitHub and ask for your data to be deleted In the event of an infringement, you may lodge a complaint with the President of the Personal Data Protection Office (uodo.gov.pl), the Polish supervisory authority.\nChanges to this policy #I reserve the right to update this policy. The date of the last change is always shown at the top of the page.\n","date":null,"permalink":"https://devlogiclabs.com/privacy/","section":"DevLogicLabs","summary":"","title":"Privacy Policy"}]