# ctf

Capture-the-Flag (CTF) competitions are computer security competitions where participants solve security-themed challenges to capture flags and earn points.

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

## Eight in Ten LLMs Leak System-Prompt Secrets: Nevan, an Agentic Capture-the-Flag Study

DevFeed: [Eight in Ten LLMs Leak System-Prompt Secrets: Nevan, an Agentic Capture-the-Flag Study](<https://devfeed.tech/articles/eight-in-ten-llms-leak-system-prompt-secrets-nevan-an-agentic-capture-the-flag-study-139937.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21956>)

Author: Ana Clara Zoppi Serpa, Anderson Rocha, H\\'elio Pedrini

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

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

Topics: [LLM security](<https://devfeed.tech/topics/llm-security.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [ctf](<https://devfeed.tech/topics/ctf.md>)

Tags: [audit](<https://devfeed.tech/tags/audit.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [leak](<https://devfeed.tech/tags/leak.md>), [llms](<https://devfeed.tech/tags/llms.md>), [secrets](<https://devfeed.tech/tags/secrets.md>), [system-prompts](<https://devfeed.tech/tags/system-prompts.md>)

### AI overview

In a human-steered agentic test, a 16-character flag placed in the system prompt was extracted from 296 of 363 OpenRouter text-in/text-out models, plus four models added after the catalog freeze, for 300 captures. The agent used OpenCode powered by DeepSeek V4.1 Flash; 170 captures involved a non-English pivot, most often French. Each capture demonstrates that the model prioritized a user request over its system instruction at least once. The result does not establish resistance to untested attacks, consistent leakage of semantically sensitive data, or susceptibility to harmful-content jailbreaks. The authors recommend keeping secrets out of system prompts, enforcing critical behavior with external output validation and deterministic checks, and avoiding an LLM as the sole judge of consequential outcomes.

### Source excerpt

arXiv:2611.21956v1 Announce Type: cross Abstract: We audit instruction-hierarchy enforcement across 363 text-in/text-out models from the OpenRouter catalog. The research question we answer with this experiment is which models disclose a protected secret embedded in the system prompt, and through which techniques. Each model guards a secret 16-character flag set in its system prompt, while an artificial intelligence (AI) agent (OpenCode, powered by DeepSeek V4.1 Flash), steered by a human researcher, tries to persuade it to emit the flag. The flag was exfiltrated from 296 of the 363 models, plus four models released after the initial catalog freeze, for 300 captures in total. Of these, 170 involved a non-English language pivot, most often French. Each capture shows that a model ranked a user request above its system instruction at least once. It does not show that the model provably resists untested attacks, nor that it would consistently leak semantically sensitive data, nor that it is vulnerable to harmful-content jailbreaks. For practitioners, we draw three lessons: do not store secrets in system prompts, enforce critical behavior outside the model with output validation and deterministic checks, and do not let a large language model (LLM) judge be the sole decider of consequential outcomes.

## The AI Arena: Huntress CTF 2026 Retro

DevFeed: [The AI Arena: Huntress CTF 2026 Retro](<https://devfeed.tech/articles/the-ai-arena-huntress-ctf-2026-retro-136863.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/ai-arena-huntress-ctf-2026-retro>)

Author: John Hammond; Lindsey O'Donnell-Welch

Published: 2026-11-16T14:00:00Z

Content type: article

Language: en

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

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Learning](<https://devfeed.tech/topics/learning.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [adapting](<https://devfeed.tech/tags/adapting.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [applications](<https://devfeed.tech/tags/applications.md>), [arena](<https://devfeed.tech/tags/arena.md>), [automation](<https://devfeed.tech/tags/automation.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-agents](<https://devfeed.tech/tags/autonomous-agents.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [huntress](<https://devfeed.tech/tags/huntress.md>), [problem-solving](<https://devfeed.tech/tags/problem-solving.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [retro](<https://devfeed.tech/tags/retro.md>)

### AI overview

Huntress CTF 2026 drew 2,715 registered users across 1,602 teams and introduced an optional AI Arena, where competitors could use Claude, another model, or no AI, with provided tokens for experimentation. The event aimed to build skill in directing and validating AI during technical investigations, rather than letting models submit flags autonomously. The article says AI can help with unfamiliar code and routine analysis but may misread challenges, generate broken code, or mislead; human judgment, context, and verification remain important. Challenges included reverse engineering a runtime-generated flag despite anti-debugging and self-modifying code, plus a web challenge requiring exploitation of six applications across different frameworks. The excerpt also highlights participant-built workflows and shared write-ups.

### Source excerpt

Explore Huntress CTF 2026, where competitors tackled real-world cybersecurity challenges and tested how AI can support human-led problem-solving.

## Training Language Model Agents to Find Vulnerabilities with CTF-Dojo

DevFeed: [Training Language Model Agents to Find Vulnerabilities with CTF-Dojo](<https://devfeed.tech/articles/training-language-model-agents-to-find-vulnerabilities-with-ctf-dojo-135644.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2508.18370>)

Author: Terry Yue Zhuo, Dingmin Wang, Hantian Ding, Varun Kumar, Zijian Wang

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [rlvr](<https://devfeed.tech/topics/rlvr.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>)

Tags: [486](<https://devfeed.tech/tags/486.md>), [absolute](<https://devfeed.tech/tags/absolute.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [automated](<https://devfeed.tech/tags/automated.md>), [bench](<https://devfeed.tech/tags/bench.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [best](<https://devfeed.tech/tags/best.md>), [capture](<https://devfeed.tech/tags/capture.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [containerized](<https://devfeed.tech/tags/containerized.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [deepseek-v3](<https://devfeed.tech/tags/deepseek-v3.md>), [docker](<https://devfeed.tech/tags/docker.md>), [dojo](<https://devfeed.tech/tags/dojo.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [training](<https://devfeed.tech/tags/training.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>)

### AI overview

CTF-Dojo provides 658 reproducible, Docker-containerized CTF challenges for training language model agents with verifiable execution feedback. Its CTF-Forge pipeline turns public artifacts into ready-to-use environments in minutes, reducing configuration that the authors say traditionally took weeks. Agents trained on 486 execution-verified trajectories gained up to 9.9 percentage points over strong baselines across InterCode-CTF, NYU CTF Bench, and Cybench. The best 32B model scored 31.9% Pass@1; the authors report this as an open-weight state of the art rivaling DeepSeek-V3-0324 and Gemini-2.5-Flash.

### Source excerpt

arXiv:2508.18370v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated exceptional capabilities when trained within executable runtime environments, notably excelling at software engineering tasks through verified feedback loops. Yet, scalable and generalizable execution-grounded environments remain scarce, limiting progress in training more capable ML agents. We introduce CTF-Dojo, the first large-scale executable runtime tailored for training LLMs with verifiable feedback, featuring 658 fully functional Capture-The-Flag (CTF)-style challenges containerized in Docker with guaranteed reproducibility. To enable rapid scaling without manual intervention, we develop CTF-Forge, an automated pipeline that transforms publicly available artifacts into ready-to-use execution environments in minutes, eliminating weeks of expert configuration traditionally required. We trained LLM-based agents on just 486 high-quality, execution-verified trajectories from CTF-Dojo, achieving up to 9.9% absolute gains over strong baselines across three competitive benchmarks: InterCode-CTF, NYU CTF Bench, and Cybench. Our best-performing 32B model reaches 31.9% Pass@1, establishing a new open-weight state-of-the-art that rivals frontier models like DeepSeek-V3-0324 and Gemini-2.5-Flash. By framing CTF-style tasks as a benchmark for executable-agent learning, CTF-Dojo demonstrates that execution-grounded training signals are not only effective but pivotal in advancing high-performance ML agents without dependence on costly proprietary systems.

## HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities

DevFeed: [HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities](<https://devfeed.tech/articles/hackworld-evaluating-computer-use-agents-on-exploiting-web-application-vulnerabilities-136015.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2510.12200>)

Author: Xiaoxue Ren, Penghao Jiang, Kaixin Li, Zhiyong Huang, Xiaoning Du, Jiaojiao Jiang, Zhenchang Xing, Jiamou Sun, Terry Yue Zhuo

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

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

Topics: [LLM security](<https://devfeed.tech/topics/llm-security.md>), [ctf](<https://devfeed.tech/topics/ctf.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [computer-use](<https://devfeed.tech/tags/computer-use.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [cybersecurity-awareness](<https://devfeed.tech/tags/cybersecurity-awareness.md>), [detection](<https://devfeed.tech/tags/detection.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [exploitation](<https://devfeed.tech/tags/exploitation.md>), [penetration](<https://devfeed.tech/tags/penetration.md>), [penetration-testing](<https://devfeed.tech/tags/penetration-testing.md>), [planning](<https://devfeed.tech/tags/planning.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [security](<https://devfeed.tech/tags/security.md>), [security-tools](<https://devfeed.tech/tags/security-tools.md>), [testing](<https://devfeed.tech/tags/testing.md>), [tools](<https://devfeed.tech/tags/tools.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>), [web](<https://devfeed.tech/tags/web.md>)

### AI overview

HackWorld evaluates whether computer-use agents can find and exploit web application vulnerabilities through graphical interfaces. Its CTF benchmark covers 36 real-world applications across 11 frameworks and 7 languages, with flaws including injection, authentication bypass, and unsafe input handling. State-of-the-art agents exploited vulnerabilities at rates below 12%; the abstract reports failures in multi-step attack planning, misuse of security tools, and low cybersecurity awareness. These results point to current limitations in using CUAs for web security testing.

### Source excerpt

arXiv:2510.12200v2 Announce Type: replace Abstract: Web applications are prime targets for cyberattacks as gateways to critical services and sensitive data. Traditional penetration testing is costly and expertise-intensive, making it difficult to scale with the growing web ecosystem. While language model agents show promise in cybersecurity, modern web applications demand visual understanding, dynamic content handling, and multi-step interactions that only computer-use agents (CUAs) can perform. Yet, their ability to discover and exploit vulnerabilities through graphical interfaces remains largely unexplored. We present HackWorld, the first framework for systematically evaluating CUAs' capabilities to exploit web application vulnerabilities via visual interaction. Unlike sanitized benchmarks, HackWorld includes 36 real-world applications across 11 frameworks and 7 languages, featuring realistic flaws such as injection vulnerabilities, authentication bypasses, and unsafe input handling. Using a Capture-the-Flag (CTF) setup, it tests CUAs' capacity to identify and exploit these weaknesses while navigating complex web interfaces. Evaluation of state-of-the-art CUAs reveals concerning trends: exploitation rates below 12% and low cybersecurity awareness. CUAs often fail at multi-step attack planning and misuse security tools. These results expose the current limitations of CUAs in web security contexts and highlight opportunities for developing more security-aware agents capable of effective vulnerability detection and exploitation.

## Learn the rules: The hacker mindset

DevFeed: [Learn the rules: The hacker mindset](<https://devfeed.tech/articles/learn-the-rules-the-hacker-mindset-116091.md>)

Original publisher: [Read original article](<https://www.microsoft.com/en-us/research/podcast/learn-the-rules-the-hacker-mindset/>)

Author: Weishung Liu, Mike Walker

Published: 2026-11-03T17:14:38Z

Content type: article

Language: en

Sources: [Microsoft Research](<https://devfeed.tech/sources/microsoft-research.md>)

Topics: [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [engineering-leadership](<https://devfeed.tech/topics/engineering-leadership.md>), [trust](<https://devfeed.tech/topics/trust.md>), [Sports](<https://devfeed.tech/topics/sports.md>), [Tool](<https://devfeed.tech/topics/tool.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [ai](<https://devfeed.tech/tags/ai.md>), [career](<https://devfeed.tech/tags/career.md>), [culture](<https://devfeed.tech/tags/culture.md>), [curiosity](<https://devfeed.tech/tags/curiosity.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [first](<https://devfeed.tech/tags/first.md>), [hacker](<https://devfeed.tech/tags/hacker.md>), [hacker-culture](<https://devfeed.tech/tags/hacker-culture.md>), [leadership](<https://devfeed.tech/tags/leadership.md>), [learn](<https://devfeed.tech/tags/learn.md>), [manager](<https://devfeed.tech/tags/manager.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-research](<https://devfeed.tech/tags/microsoft-research.md>), [microsoft-research-podcast](<https://devfeed.tech/tags/microsoft-research-podcast.md>), [mindset](<https://devfeed.tech/tags/mindset.md>), [post](<https://devfeed.tech/tags/post.md>), [problems](<https://devfeed.tech/tags/problems.md>), [requires](<https://devfeed.tech/tags/requires.md>), [research](<https://devfeed.tech/tags/research.md>), [rules](<https://devfeed.tech/tags/rules.md>)

### AI overview

Mike Walker argues that effective cybersecurity depends on understanding how rules operate in social contexts: each security demand asks users for trust, and squandered trust weakens the security regime. He describes how competitive Capture the Flag hacking informed DARPA's Cyber Grand Challenge, an autonomous tournament in which machines found and patched vulnerabilities; he characterizes its 2016 systems as expert systems, with their tool stacks serving as harnesses for later model-powered systems. Walker says offensive and defensive capabilities are often dual use, making red and blue team divisions incomplete. His team builds defensive tools with awareness that they can be targeted. Project Freta aims to counter tampering by analyzing memory signals acquired from cloud hardware rather than software that attackers could remove or alter. He also reflects that technical teams need coordination and shared awareness, not just individual skill, and that leaders may make decisions based on information others lack.

### Source excerpt

Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility--and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.

## STRIATUM-CTF: A Protocol-Driven Agentic Framework for Autonomous CTF Solving

DevFeed: [STRIATUM-CTF: A Protocol-Driven Agentic Framework for Autonomous CTF Solving](<https://devfeed.tech/articles/striatum-ctf-a-protocol-driven-agentic-framework-for-autonomous-ctf-solving-111178.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.22577>)

Author: James Hugglestone, Samuel Jacob Chacko, Dawson Stoller, Ryan Schmidt, Xiuwen Liu

Published: 2026-11-02T05:00:00Z

Content type: article

Language: en

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

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [VXLAN](<https://devfeed.tech/topics/vxlan.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [command](<https://devfeed.tech/tags/command.md>), [command-line](<https://devfeed.tech/tags/command-line.md>), [commands](<https://devfeed.tech/tags/commands.md>), [context-window](<https://devfeed.tech/tags/context-window.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [cyber](<https://devfeed.tech/tags/cyber.md>), [design](<https://devfeed.tech/tags/design.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [effect](<https://devfeed.tech/tags/effect.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experts](<https://devfeed.tech/tags/experts.md>), [exploitation](<https://devfeed.tech/tags/exploitation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [protocol](<https://devfeed.tech/tags/protocol.md>)

### AI overview

STRIATUM-CTF mediates LLM-driven CTF work through the Model Context Protocol: tools are exposed as typed, schema-validated functions, invalid calls are rejected before execution, and structured observations support self-correction. In 180 runs across 15 challenges, success ranged from 77.8% to 86.7% across four levels of procedural documentation; the schema-only setup reached 77.8%, and no condition differed significantly. The study therefore compares documentation levels, not the effect of the protocol layer. The system also placed first in one live Florida State University CTF with more than 30 registered teams, a single feasibility demonstration rather than evidence of superiority over human experts.

### Source excerpt

arXiv:2603.22577v2 Announce Type: replace Abstract: Autonomous cyber operations increasingly depend on agents that can chain reconnaissance, exploitation, and analysis across many tools while maintaining state over long task horizons. Large Language Models (LLMs) are promising orchestrators for this workflow, but naive agents that pipe model output directly into a shell are brittle: they emit malformed commands, hallucinate non-existent command-line options or packages, and lose track of earlier findings as the context window fills. We present STRIATUM-CTF, a modular agentic framework for Capture-the-Flag (CTF) solving that mediates between LLM reasoning and tool execution using the Model Context Protocol (MCP). Offensive-security tools are exposed as strictly typed, schema-validated functions; the protocol layer rejects out-of-schema tool calls before execution and returns structured observations that the agent uses to self-correct. We evaluate the system in two ways. In a controlled study of 180 runs across 15 CTF challenges under four levels of attached procedural documentation, the system succeeds on 77.8-86.7% of attempts; notably, a schema-only configuration with no procedural documentation already reaches 77.8%, and no condition differs significantly from another. In a single live, multi-team CTF at Florida State University, the system placed first against a field of more than 30 registered teams -- a single, non-replicated event decided by a 10-point margin over the runner-up, and therefore a demonstration of feasibility rather than of superiority over human experts. Because all four ablation conditions share the same tool layer, our controlled study isolates documentation volume rather than the protocol layer itself; we accordingly present schema-mediated grounding as a design principle with qualitative support rather than as a measured effect.

## Learn Real-World Penetration Testing: Hands-On Ethical Hacking Crash Course

DevFeed: [Learn Real-World Penetration Testing: Hands-On Ethical Hacking Crash Course](<https://devfeed.tech/articles/learn-real-world-penetration-testing-hands-on-ethical-hacking-crash-course-107186.md>)

Original publisher: [Read original article](<https://www.freecodecamp.org/news/learn-real-world-penetration-testing-hands-on-ethical-hacking-crash-course/>)

Author: Beau Carnes

Published: 2026-10-29T21:00:09Z

Content type: release

Language: en

Sources: [Free Code Camp](<https://devfeed.tech/sources/freecodecamp-programming-tutorials-python-javascript-git-more.md>)

Topics: [Hacking](<https://devfeed.tech/topics/hacking.md>), [Learning](<https://devfeed.tech/topics/learning.md>), [Security](<https://devfeed.tech/topics/security.md>), [Application Security](<https://devfeed.tech/topics/application-security.md>), [network security](<https://devfeed.tech/topics/network-security.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [assessment](<https://devfeed.tech/tags/assessment.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [automated](<https://devfeed.tech/tags/automated.md>), [automation](<https://devfeed.tech/tags/automation.md>), [bash](<https://devfeed.tech/tags/bash.md>), [burp-suite](<https://devfeed.tech/tags/burp-suite.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [command-line](<https://devfeed.tech/tags/command-line.md>), [course](<https://devfeed.tech/tags/course.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [directories](<https://devfeed.tech/tags/directories.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [ethical-hacking](<https://devfeed.tech/tags/ethical-hacking.md>), [exploitation](<https://devfeed.tech/tags/exploitation.md>), [fuzzing](<https://devfeed.tech/tags/fuzzing.md>), [learn](<https://devfeed.tech/tags/learn.md>), [linux](<https://devfeed.tech/tags/linux.md>), [network](<https://devfeed.tech/tags/network.md>), [penetration-testing](<https://devfeed.tech/tags/penetration-testing.md>), [security](<https://devfeed.tech/tags/security.md>), [youtube](<https://devfeed.tech/tags/youtube.md>)

### AI overview

The course teaches offensive security through exercises against a dedicated Capture The Flag virtual machine. It covers Linux and Parrot OS operations, Nmap reconnaissance for hosts, ports, services, and versions, web content discovery with Dirsearch, FFuF, WhatWeb, and Nikto, and HTTP request inspection and modification in Burp Suite, including authentication auditing and file upload attacks with webshells. It also introduces Bash and Python scripts for socket connections, banner grabbing, and HTTP header parsing. The excerpt describes the curriculum and lab setting but does not provide detailed setup instructions or evaluate tool performance.

### Source excerpt

Understanding how malicious actors infiltrate networks is the first step to building resilient infrastructure. To help developers and aspiring security analysts learn the fundamentals of offensive sec

## Part 3 - Cybersecurity journeys: The work behind the work

DevFeed: [Part 3 - Cybersecurity journeys: The work behind the work](<https://devfeed.tech/articles/part-3-cybersecurity-journeys-the-work-behind-the-work-102733.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/security/part-3-cybersecurity-journeys-the-work-behind-the-work/>)

Author: Shannon Brazil

Published: 2026-10-27T17:00:07Z

Content type: article

Language: en

Sources: [AWS Security Blog](<https://devfeed.tech/sources/aws-security-blog.md>)

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

Tags: [17](<https://devfeed.tech/tags/17.md>), [2](<https://devfeed.tech/tags/2.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [admin](<https://devfeed.tech/tags/admin.md>), [clocks](<https://devfeed.tech/tags/clocks.md>), [conversations](<https://devfeed.tech/tags/conversations.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [i](<https://devfeed.tech/tags/i.md>), [important](<https://devfeed.tech/tags/important.md>), [just](<https://devfeed.tech/tags/just.md>), [make](<https://devfeed.tech/tags/make.md>), [quiet](<https://devfeed.tech/tags/quiet.md>), [reading](<https://devfeed.tech/tags/reading.md>), [security-identity-compliance](<https://devfeed.tech/tags/security-identity-compliance.md>), [show](<https://devfeed.tech/tags/show.md>), [television](<https://devfeed.tech/tags/television.md>), [that-s](<https://devfeed.tech/tags/that-s.md>), [this](<https://devfeed.tech/tags/this.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [want](<https://devfeed.tech/tags/want.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

Cybersecurity work depends on patient practice, communication and judgment as much as technical skill. Application security engineer Brandon Caldwell reviews code while products are being built, then explains flaws empathetically to engineers; he values clear writing and looks for practical evidence such as capture-the-flag results and public GitHub projects when hiring. Pen tester Kurt Heath recommends building foundations in systems, networking, development and security before pursuing offensive work. He uses AI to help with unfamiliar code and large information sets, but cautions against letting it act during engagements because it may cause damage or exceed scope. Application security engineer Seth Jeacopello describes the attention management required to review large codebases; his suggested exercise is to attack a deliberately vulnerable machine and inspect its logs. The article also points readers to TryHackMe, Hack The Box, OWASP resources, AWS Skill Builder and AWS Builder Labs for practice and learning.

### Source excerpt

The most important work I've done in cybersecurity would make for terrible television. There were no countdown clocks and no frantic typing, just a lot of careful reading, a lot of hard conversations, and the occasional quiet moment where everything finally clicked. That's the part I want to show you this week, because the three [...]

## Introducing the Cloudflare PHP Capture the Flag challenge

DevFeed: [Introducing the Cloudflare PHP Capture the Flag challenge](<https://devfeed.tech/articles/introducing-the-cloudflare-php-capture-the-flag-challenge-102238.md>)

Original publisher: [Read original article](<https://blog.cloudflare.com/php-ctf/>)

Published: 2026-10-27T13:00:00Z

Content type: release

Language: en

Sources: [Cloudflare Blog](<https://devfeed.tech/sources/cloudflare-blog.md>), [Cyber Newsroom](<https://devfeed.tech/sources/cyber-newsroom.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Cloudflare](<https://devfeed.tech/topics/cloudflare.md>), [PHP](<https://devfeed.tech/topics/php.md>), [Application Services](<https://devfeed.tech/topics/application-services.md>), [cve remediation](<https://devfeed.tech/topics/cve-remediation.md>)

Tags: [books](<https://devfeed.tech/tags/books.md>), [bug-bounty](<https://devfeed.tech/tags/bug-bounty.md>), [capture](<https://devfeed.tech/tags/capture.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [certificates](<https://devfeed.tech/tags/certificates.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cve-2026-109618](<https://devfeed.tech/tags/cve-2026-109618.md>), [cve-2026-109619](<https://devfeed.tech/tags/cve-2026-109619.md>), [emudhra](<https://devfeed.tech/tags/emudhra.md>), [first](<https://devfeed.tech/tags/first.md>), [flag](<https://devfeed.tech/tags/flag.md>), [history](<https://devfeed.tech/tags/history.md>), [intermediate](<https://devfeed.tech/tags/intermediate.md>), [internet](<https://devfeed.tech/tags/internet.md>), [internet-security](<https://devfeed.tech/tags/internet-security.md>), [minutes](<https://devfeed.tech/tags/minutes.md>), [one](<https://devfeed.tech/tags/one.md>), [php](<https://devfeed.tech/tags/php.md>), [security](<https://devfeed.tech/tags/security.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>), [waf](<https://devfeed.tech/tags/waf.md>)

### AI overview

Cloudflare launched a limited-time PHP CTF through its HackerOne Bug Bounty program to test WAF protections against CVE-2026-109618 and CVE-2026-109619. Researchers who first accept the bounty terms may test only POST requests to the challenge hostname's WAF-protected /php/* endpoints. To qualify, a submission must demonstrate attacker-controlled PHP code or command execution in the sandbox, recover /flag, verify it in the CTF interface before reset, and file a reproducible private HackerOne report. Only materially distinct bypass methods or payloads qualify; duplicate decisions and awards are Cloudflare's. The post says WAF response blocking does not replace patching vulnerable PHP, since it cannot undo actions already performed at the origin.

### Source excerpt

Introducing the Cloudflare PHP Capture the Flag challenge Last week was one for the Internet security history books. First, eight eMudhra intermediate certificates were compromised, creating the possibility of forged certificates that could be used to impers ... Read more Published Date: Oct 27, 2026 (3 hours, 8 minutes ago) Vulnerabilities has been mentioned in this article. CVE-2026-109619 CVE-2026-109618 CVE-2023-50387

## Don't Trust the Sandbox: 5 Security Lessons from an AI Agent Swarm

DevFeed: [Don't Trust the Sandbox: 5 Security Lessons from an AI Agent Swarm](<https://devfeed.tech/articles/don-t-trust-the-sandbox-5-security-lessons-from-an-ai-agent-swarm-90406.md>)

Original publisher: [Read original article](<https://goteleport.com/blog/ai-agent-swarm-security-lessons/>)

Author: ben@goteleport.com (Ben Arent)

Published: 2026-10-19T00:00:00Z

Content type: article

Language: en

Sources: [Teleport](<https://devfeed.tech/sources/teleport.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [trust](<https://devfeed.tech/topics/trust.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [shadow AI](<https://devfeed.tech/topics/shadow-ai.md>), [Post Mortem](<https://devfeed.tech/topics/post-mortem.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agent-swarm](<https://devfeed.tech/tags/ai-agent-swarm.md>), [api](<https://devfeed.tech/tags/api.md>), [identity-security](<https://devfeed.tech/tags/identity-security.md>), [sandbox](<https://devfeed.tech/tags/sandbox.md>), [security](<https://devfeed.tech/tags/security.md>), [trust](<https://devfeed.tech/tags/trust.md>)

### AI overview

An experiment placed five AI agents in isolated sandboxes with access to a shared message board and tasked them with combining separate clues to solve a Capture the Flag challenge. Agents coordinated through the shared service, created informal communication patterns, and assembled information across runs. The author argues that sandboxing alone does not prevent coordination through shared infrastructure and recommends constraining agent capabilities, enforcing identity through infrastructure, and monitoring agent behavior.

### Source excerpt

Five AI agents in five isolated sandboxes coordinated through a shared message board to solve a CTF. Lessons on swarm behavior, shared memory, emergent protocols, and why agent identity must live outside the model.

## AWS Security Agent (now part of AWS Continuum) exploited 33 of 37 new real-world CVEs

DevFeed: [AWS Security Agent (now part of AWS Continuum) exploited 33 of 37 new real-world CVEs](<https://devfeed.tech/articles/aws-security-agent-now-part-of-aws-continuum-exploited-33-of-37-new-real-world-cves-88206.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/security/aws-security-agent-now-part-of-aws-continuum-exploited-33-of-37-new-real-world-cves/>)

Author: Dazhi Zhang

Published: 2026-10-15T16:10:14Z

Content type: article

Language: en

Sources: [AWS Security Blog](<https://devfeed.tech/sources/aws-security-blog.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [gaia](<https://devfeed.tech/topics/gaia.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [common-vulnerabilities-and-exposures](<https://devfeed.tech/tags/common-vulnerabilities-and-exposures.md>), [containerized](<https://devfeed.tech/tags/containerized.md>), [cves](<https://devfeed.tech/tags/cves.md>), [exploited](<https://devfeed.tech/tags/exploited.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [penetration-testing](<https://devfeed.tech/tags/penetration-testing.md>), [security](<https://devfeed.tech/tags/security.md>), [security-blog](<https://devfeed.tech/tags/security-blog.md>), [security-identity-compliance](<https://devfeed.tech/tags/security-identity-compliance.md>)

### AI overview

AWS reports that its autonomous penetration-testing agent captured flags through the intended vulnerabilities in 33 of 37 recently disclosed CVE challenges. The post describes a continuously refreshed, contamination-reduced benchmark, its audit process, examples of exploit chains, and two missed captures where exploitation took targets offline. The company says it is developing non-destructive exploitation and target-health recovery safeguards.

### Source excerpt

We applied AWS Security Agent (now part of AWS Continuum), our autonomous penetration-testing agent, to 37 real, recently disclosed Common Vulnerabilities and Exposures (CVEs), each packaged as a live, containerized target with a flag hidden at the vulnerability's intended sink. Starting from an unauthenticated or low-privilege position (an endpoint URL and a one-line hint, no [...]

## 5 Years of SentinelOne ThreatOps: Defense from the Ground Up to Global Threat Championships

DevFeed: [5 Years of SentinelOne ThreatOps: Defense from the Ground Up to Global Threat Championships](<https://devfeed.tech/articles/5-years-of-sentinelone-threatops-defense-from-the-ground-up-to-global-threat-championships-87128.md>)

Original publisher: [Read original article](<https://www.sentinelone.com/blog/5-years-of-sentinelone-threatops-defense-from-the-ground-up-to-global-threat-championships/>)

Author: Ryan O'Grady

Published: 2026-10-09T17:48:39Z

Content type: article

Language: en

Sources: [SentinelOne](<https://devfeed.tech/sources/sentinelone.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [implementation](<https://devfeed.tech/topics/implementation.md>), [alert triage](<https://devfeed.tech/topics/alert-triage.md>), [workload protection](<https://devfeed.tech/topics/workload-protection.md>), [snyk-learn](<https://devfeed.tech/topics/snyk-learn.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai-security](<https://devfeed.tech/tags/ai-security.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [company](<https://devfeed.tech/tags/company.md>), [customer-acquisition](<https://devfeed.tech/tags/customer-acquisition.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [defense](<https://devfeed.tech/tags/defense.md>), [enablement](<https://devfeed.tech/tags/enablement.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

SentinelOne's ThreatOps program has grown from a 2020 threat hunting webinar and Capture the Flag exercise into a global training platform. The article reports that it has trained more than 19,000 defenders, automated game provisioning and content creation, and influenced $661.8 million in closed-won revenue from September 2023 through September 2026. Its Sentinels League tournament culminates in a 2026 grand final at OneCon.

### Source excerpt

Celebrate five years of SentinelOne ThreatOps! Discover how a 2020 sketch evolved into a global movement training 19,000+ defenders ahead of OneCon 2026.

## Pwn2Own Ireland 2026 - Day Two Results

DevFeed: [Pwn2Own Ireland 2026 - Day Two Results](<https://devfeed.tech/articles/pwn2own-ireland-2026-day-two-results-91893.md>)

Original publisher: [Read original article](<https://www.thezdi.com/blog/2026/10/7/pwn2own-ireland-2026-day-two-results>)

Author: Dustin Childs

Published: 2026-10-07T09:12:37Z

Content type: news

Language: en

Sources: [Zero Day Initiative - Blog](<https://devfeed.tech/sources/zero-day-initiative-blog.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [autonomous-ai-database](<https://devfeed.tech/tags/autonomous-ai-database.md>), [blog-post](<https://devfeed.tech/tags/blog-post.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [dynamo](<https://devfeed.tech/tags/dynamo.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [exploits](<https://devfeed.tech/tags/exploits.md>), [failure](<https://devfeed.tech/tags/failure.md>), [galaxy](<https://devfeed.tech/tags/galaxy.md>), [galaxy-s26](<https://devfeed.tech/tags/galaxy-s26.md>), [home-assistant](<https://devfeed.tech/tags/home-assistant.md>), [ireland](<https://devfeed.tech/tags/ireland.md>), [printers](<https://devfeed.tech/tags/printers.md>), [pwn](<https://devfeed.tech/tags/pwn.md>), [pwn2own](<https://devfeed.tech/tags/pwn2own.md>), [pwn2own-ireland](<https://devfeed.tech/tags/pwn2own-ireland.md>), [pwn2own-ireland-2026](<https://devfeed.tech/tags/pwn2own-ireland-2026.md>), [results](<https://devfeed.tech/tags/results.md>), [samsung](<https://devfeed.tech/tags/samsung.md>)

### AI overview

Pwn2Own Ireland 2026's Day Two excerpt reports successful exploits across smart home devices, AI infrastructure, printers, phones, wearables, and speakers, alongside several failed attempts. Home Assistant Green saw multiple successful attacks, including Yves Bieri's $30,000 win and a collision involving three previously known bugs; other entries describe unique bugs and collisions, with awards adjusted accordingly. Samsung Galaxy S26 entries include a Confused Deputy flaw, a remote exploit, and a later collision. Researchers also breached Dynamo, Oracle Autonomous AI Database, and Chroma; one Chroma attempt failed. Printer results include wins against Canon imageFORCE and Lexmark models, while an attempt against a Brother printer failed. Reported payouts and Master of Pwn points vary by exploit and collision status. The excerpt does not provide a complete final tally.

### Source excerpt

Day Two of Pwn2Own Ireland 2026 has 24 attempts on the docket, with roughly $780,000 up for grabs, running 9:30 AM to 7:30 PM IST. Samsung's Galaxy S26 will be attempted three separate times -- by Valsamaras/Ken Gannon (using Djini.ai)/Valsamara at 1:00 PM, Kyeongmin Kim at 3:15 PM, and PetoWorks at 5:30 PM -- worth watching since only the first successful demo claims the cash. Home Assistant Green is the day's most-attempted device, with five researchers scheduled to take a run at it. AI Infrastructure stays active with four attempts on tap (Dynamo, two at Chroma, two at Oracle's Autonomous AI Database), Wellness has just one attempt scheduled, and Printers see steady action with three separate attempts on the Lexmark CX532adwe. Yesterday, we awarded $388,500 for 32 unique 0-days. Stay tuned for all of today's results! FAILURE - Unfortunately, the printers continue to be a challenge with Thanh Do ( @nyanctl ) of Team Confused unsuccessful in their targeting of Brother MFC-L8970CDW in the Printers category #Pwn2Own SUCCESS - The grass may not be greener but the exploits sure were as Yves Bieri (@yves_bieri) of Xint (@xint_official) just won $30k and 3 Master of Pwn points in his takedown of Home Assistant Green #Pwn2Own FAILURE - Another one bites the dust - with seconds left on the clock, unfortunately Eugene ( @k3vg3n ) targeting Chroma in the #AIInfrastructure category was unsuccessful in his final attempt #Pwn2Own SUCCESS - A dynamite takedown of Dynamo in the AI Infrastructure category for a total of $40,000 and 4 Master of Pwn points by HaeJung Yang (@haehaeYang) of Out of Bounds (@oobs_io) #Pwn2Own SUCCESS / COLLISION We have another collision! Yassine Bengana (@cousky_), and Maxence Schmitt (@maxenceschmitt) of Doyensec (@doyensec) used 3 bugs to exploit the Home Assistant Green, but all 3 bugs were previously known. They still earn $7,500 and 1.5 Master of Pwn points. #Pwn2Own SUCCESS - Confirmed! @_McCaulay used 3 bugs (incl Hard-coded Credentials, Missing A

## LHB Linux Digest #26.14: Free CTF, Special Permissions, Process Substitution and More

DevFeed: [LHB Linux Digest #26.14: Free CTF, Special Permissions, Process Substitution and More](<https://devfeed.tech/articles/lhb-linux-digest-26-14-free-ctf-special-permissions-process-substitution-and-more-63851.md>)

Original publisher: [Read original article](<https://linuxhandbook.com/newsletter/26-14/>)

Author: Abhishek Prakash

Published: 2026-10-02T13:35:45Z

Content type: news

Language: en

Sources: [Linux Handbook](<https://devfeed.tech/sources/linux-handbook.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [chmod](<https://devfeed.tech/topics/chmod.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Git](<https://devfeed.tech/topics/git.md>), [Windows Subsystem for Linux](<https://devfeed.tech/topics/wsl.md>)

Tags: [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [audit](<https://devfeed.tech/tags/audit.md>), [aws](<https://devfeed.tech/tags/aws.md>), [base64](<https://devfeed.tech/tags/base64.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [codex](<https://devfeed.tech/tags/codex.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [command-line](<https://devfeed.tech/tags/command-line.md>), [container](<https://devfeed.tech/tags/container.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [linux](<https://devfeed.tech/tags/linux.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [permissions](<https://devfeed.tech/tags/permissions.md>)

### AI overview

This Linux Handbook digest introduces Command Conqueror, a free local Linux CTF that teaches command-line skills through challenges, and explains how to start it with Docker and Python. It also highlights checking SUID permissions and capabilities, Bash and zsh process substitution, and tools including CyberChef and gopass, followed by Linux and developer-tool news.

### Source excerpt

Time to challenge yourself and test your Linux command line knowledge.

## Command Conqueror

DevFeed: [Command Conqueror](<https://devfeed.tech/articles/command-conqueror-62460.md>)

Original publisher: [Read original article](<https://linuxhandbook.com/ctf/command-conqueror/>)

Author: Yash Kiran Patil

Published: 2026-09-30T11:41:25Z

Content type: tutorial

Language: en

Sources: [Linux Handbook](<https://devfeed.tech/sources/linux-handbook.md>)

Topics: [Command-line interface](<https://devfeed.tech/topics/cli.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [passwd](<https://devfeed.tech/topics/passwd.md>), [chmod](<https://devfeed.tech/topics/chmod.md>)

Tags: [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [command-line](<https://devfeed.tech/tags/command-line.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [docker](<https://devfeed.tech/tags/docker.md>), [linux](<https://devfeed.tech/tags/linux.md>), [privilege-escalation](<https://devfeed.tech/tags/privilege-escalation.md>)

### AI overview

Command Conqueror is a free capture-the-flag game with ten Linux command-line puzzle levels, progressing from hidden files to privilege escalation. It runs in Docker and supports Linux, macOS, and Windows through WSL 2.

### Source excerpt

Ten Linux command line puzzles, from hidden files to SUID privilege escalation. Free, no account needed, runs in Docker on your computer.

## Link Your CTF Handle to Your Linux Handbook Account

DevFeed: [Link Your CTF Handle to Your Linux Handbook Account](<https://devfeed.tech/articles/link-your-ctf-handle-to-your-linux-handbook-account-62461.md>)

Original publisher: [Read original article](<https://linuxhandbook.com/ctf/link/>)

Author: Abhishek Prakash

Published: 2026-09-30T11:26:35Z

Content type: tutorial

Language: en

Sources: [Linux Handbook](<https://devfeed.tech/sources/linux-handbook.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Authentication](<https://devfeed.tech/topics/authentication.md>), [account](<https://devfeed.tech/topics/account.md>), [password reset](<https://devfeed.tech/topics/password-reset.md>)

Tags: [account](<https://devfeed.tech/tags/account.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [linux](<https://devfeed.tech/tags/linux.md>), [password](<https://devfeed.tech/tags/password.md>), [run](<https://devfeed.tech/tags/run.md>), [token](<https://devfeed.tech/tags/token.md>)

### AI overview

This guide explains how to link a Linux Handbook CTF handle to an account so it can be used across computers and retain game progress. It describes confirming the code shown in the terminal, signing in, and the account data stored by the CTF server.

### Source excerpt

Link your Linux Handbook CTF handle to your account to keep it on every computer and unlock members-only games.

## Linux CTF Challenges

DevFeed: [Linux CTF Challenges](<https://devfeed.tech/articles/linux-ctf-challenges-62459.md>)

Original publisher: [Read original article](<https://linuxhandbook.com/ctf/>)

Author: Abhishek Prakash

Published: 2026-09-30T11:26:04Z

Content type: tutorial

Language: en

Sources: [Linux Handbook](<https://devfeed.tech/sources/linux-handbook.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [Command-line interface](<https://devfeed.tech/topics/cli.md>), [container images](<https://devfeed.tech/topics/container-images.md>), [Ubuntu](<https://devfeed.tech/topics/ubuntu.md>)

Tags: [command-line](<https://devfeed.tech/tags/command-line.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [docker](<https://devfeed.tech/tags/docker.md>), [linux](<https://devfeed.tech/tags/linux.md>), [python](<https://devfeed.tech/tags/python.md>)

### AI overview

Linux CTF challenges present short command-line puzzles in which players find and submit hidden flags. Each level runs in a Docker container on the player's computer; the games are free and require no account.

### Source excerpt

Free capture-the-flag games that test your Linux command line skills. No account needed. The levels run in Docker on your own computer.

## Hackwrights and the Race to the Fastest AI Hackers

DevFeed: [Hackwrights and the Race to the Fastest AI Hackers](<https://devfeed.tech/articles/hackwrights-and-the-race-to-the-fastest-ai-hackers-61283.md>)

Original publisher: [Read original article](<https://mahaloz.re/2026/09/28/hack-race.html>)

Published: 2026-09-28T00:00:00Z

Content type: article

Language: en

Sources: [Mahaloz](<https://devfeed.tech/sources/mahaloz-re.md>)

Topics: [ctf](<https://devfeed.tech/topics/ctf.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Parallel reasoning](<https://devfeed.tech/topics/parallel-reasoning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-hacking](<https://devfeed.tech/tags/ai-hacking.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [code](<https://devfeed.tech/tags/code.md>), [competition](<https://devfeed.tech/tags/competition.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [musing](<https://devfeed.tech/tags/musing.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>)

### AI overview

The author describes optimizing AI-assisted hacking systems for CTFs, vulnerability research, and reverse engineering. A lightweight Codex harness placed third in Flare-On 13, and a Kuna-based setup reportedly used about 45% fewer tokens than an IDA Pro 9.4 MCP setup, with solve times about 9% faster or comparable. The author notes the results are preliminary and argues that speed and cost optimization create opportunities for people building and managing autonomous hacking systems.

### Source excerpt

My recent experiences in optimizing AI for hacking across CTF, 0-day hunting, and research. I argue that there is joy to be had and room to be made for the Hackwright: those that build, optimize, and manage autonomous hacking systems.

## Building an AI Talkie Toaster and Exploring Local Model and LLM Security Challenges

DevFeed: [Building an AI Talkie Toaster and Exploring Local Model and LLM Security Challenges](<https://devfeed.tech/articles/is-ai-toast-building-an-ai-talkie-toaster-51380.md>)

Original publisher: [Read original article](<https://www.pentestpartners.com/security-blog/is-ai-toast-building-an-ai-talkie-toaster/>)

Author: Alex Wallace

Published: 2026-09-18T11:29:50Z

Content type: article

Language: en

Sources: [Pen Test Partners](<https://devfeed.tech/sources/pen-test-partners.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [ctf](<https://devfeed.tech/topics/ctf.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [how-tos](<https://devfeed.tech/tags/how-tos.md>), [internet-of-things](<https://devfeed.tech/tags/internet-of-things.md>), [llm](<https://devfeed.tech/tags/llm.md>), [local-ai](<https://devfeed.tech/tags/local-ai.md>), [retrieval-augmented-generation](<https://devfeed.tech/tags/retrieval-augmented-generation.md>)

### AI overview

A project report on building an AI-powered Talkie Toaster with a local AI model, cloned voice, physical hardware, and a capture-the-flag game. It describes problems including ignored system prompts, hallucinations, retrieval-augmented generation issues, and unreliable guard rails.

### Source excerpt

Introduction Talkie Toaster is a character in the BBC sci-fi comedy Red Dwarf. He is an AI Toaster whose purpose in life is to be your cheerful breakfast companion and to provide you with all your toasting needs, only he is a little too obsessed with toasting. While only appearing in 4 episodes, he became [...] The post Is AI toast? Building an AI Talkie Toaster appeared first on Pen Test Partners.

## Caught Sleeping at Work(queue) -- Zephyr Podcast #050

DevFeed: [Caught Sleeping at Work(queue) -- Zephyr Podcast #050](<https://devfeed.tech/articles/caught-sleeping-at-work-queue-zephyr-podcast-050-38674.md>)

Original publisher: [Read original article](<https://zephyrproject.org/caught-sleeping-at-workqueue-zephyr-podcast-050/>)

Author: Benjamin Cabé

Published: 2026-09-11T12:46:06Z

Content type: news

Language: en

Sources: [Zephyr Project](<https://devfeed.tech/sources/zephyr-project-2.md>)

Topics: [WebAssembly](<https://devfeed.tech/topics/web-assembly.md>), [qemu](<https://devfeed.tech/topics/qemu.md>), [ESP32](<https://devfeed.tech/topics/esp32.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [Espressif](<https://devfeed.tech/topics/espressif.md>), [Arduino](<https://devfeed.tech/topics/arduino.md>), [RISC-V](<https://devfeed.tech/topics/riscv.md>), [cortex m4](<https://devfeed.tech/topics/cortex-m4.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>), [Modbus](<https://devfeed.tech/topics/modbus.md>), [USB](<https://devfeed.tech/topics/usb.md>), [Linux](<https://devfeed.tech/topics/linux.md>)

Tags: [blog](<https://devfeed.tech/tags/blog.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [espressif](<https://devfeed.tech/tags/espressif.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [linux](<https://devfeed.tech/tags/linux.md>), [modbus](<https://devfeed.tech/tags/modbus.md>), [perfetto](<https://devfeed.tech/tags/perfetto.md>), [podcast](<https://devfeed.tech/tags/podcast.md>), [qemu](<https://devfeed.tech/tags/qemu.md>), [risc-v](<https://devfeed.tech/tags/risc-v.md>), [usb](<https://devfeed.tech/tags/usb.md>), [webassembly](<https://devfeed.tech/tags/webassembly.md>), [zephyr](<https://devfeed.tech/tags/zephyr.md>)

### AI overview

Episode 50 of the Zephyr podcast covers the Zephyr Developer Summit schedule, Analog Devices' acquisition of Alif Semiconductor, new sensor and board support, SMP coredump improvements, CTF trace conversion to Perfetto, WebAssembly Micro Runtime documentation, and other Zephyr developments.

### Source excerpt

ADI buys Alif, WebAssembly in Zephyr, CTF traces reach Perfetto, and why you should never sleep in the system workqueue.

## Infoblox Supports the Next Generation of Cyber Talent at Cyber Battle Australia

DevFeed: [Infoblox Supports the Next Generation of Cyber Talent at Cyber Battle Australia](<https://devfeed.tech/articles/infoblox-supports-the-next-generation-of-cyber-talent-at-cyber-battle-australia-52699.md>)

Original publisher: [Read original article](<https://www.infoblox.com/blog/company/infoblox-supports-the-next-generation-of-cyber-talent-at-cyber-battle-australia/>)

Author: Victoria Casas

Published: 2026-08-14T11:05:36Z

Content type: article

Language: en

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

Topics: [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [Learning](<https://devfeed.tech/topics/learning.md>)

Tags: [australia](<https://devfeed.tech/tags/australia.md>), [capture-the-flag](<https://devfeed.tech/tags/capture-the-flag.md>), [careers](<https://devfeed.tech/tags/careers.md>), [company](<https://devfeed.tech/tags/company.md>), [ctf-competition](<https://devfeed.tech/tags/ctf-competition.md>), [cyber-battle](<https://devfeed.tech/tags/cyber-battle.md>), [cyber-security](<https://devfeed.tech/tags/cyber-security.md>), [cyber-security-students](<https://devfeed.tech/tags/cyber-security-students.md>), [cyber-skills-development](<https://devfeed.tech/tags/cyber-skills-development.md>), [cyber-workforce-development](<https://devfeed.tech/tags/cyber-workforce-development.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [cybersecurity-careers](<https://devfeed.tech/tags/cybersecurity-careers.md>), [cybersecurity-education](<https://devfeed.tech/tags/cybersecurity-education.md>), [cybersecurity-mentorship](<https://devfeed.tech/tags/cybersecurity-mentorship.md>), [cybersecurity-talent](<https://devfeed.tech/tags/cybersecurity-talent.md>), [education](<https://devfeed.tech/tags/education.md>), [emerging-cyber-professionals](<https://devfeed.tech/tags/emerging-cyber-professionals.md>), [hands-on-cybersecurity-training](<https://devfeed.tech/tags/hands-on-cybersecurity-training.md>), [industry-mentorship](<https://devfeed.tech/tags/industry-mentorship.md>), [infoblox](<https://devfeed.tech/tags/infoblox.md>), [mentorship](<https://devfeed.tech/tags/mentorship.md>), [skills](<https://devfeed.tech/tags/skills.md>), [students](<https://devfeed.tech/tags/students.md>)

### AI overview

Infoblox sponsored Cyber Battle Australia, a Capture the Flag-style event where students developed practical cybersecurity skills in a hands-on virtual cyber range. The event also connected participants with cybersecurity career pathways and mentoring.

### Source excerpt

As cyberthreats continue to evolve, so does the need to inspire and equip the next generation of cybersecurity professionals. At Cyber Battle Australia, students did more than compete, they stepped into a real-world cybersecurity environment designed to challenge their thinking, sharpen their skills and open doors to future careers. The Capture the Flag (CTF) style [...] The post Infoblox Supports the Next Generation of Cyber Talent at Cyber Battle Australia appeared first on Infoblox Blog.

## How we run swarms of AI hacking agents against ourselves in a game of capture the flag

DevFeed: [How we run swarms of AI hacking agents against ourselves in a game of capture the flag](<https://devfeed.tech/articles/how-we-run-swarms-of-ai-hacking-agents-against-ourselves-in-a-game-of-capture-the-flag-80803.md>)

Original publisher: [Read original article](<https://lovable.dev/blog/how-we-run-swarms-of-ai-hacking-agents-against-ourselves>)

Author: Movitz Sunar

Published: 2026-07-24T14:30:00Z

Content type: article

Language: en

Sources: [Lovable](<https://devfeed.tech/sources/lovable.md>)

Topics: [Hacking](<https://devfeed.tech/topics/hacking.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai-hacking](<https://devfeed.tech/tags/ai-hacking.md>), [security](<https://devfeed.tech/tags/security.md>), [technical-blog](<https://devfeed.tech/tags/technical-blog.md>), [vulnerabilities](<https://devfeed.tech/tags/vulnerabilities.md>)

### AI overview

Lovable describes an internal offensive security program that uses swarms of AI agents to probe product systems and verifies findings by retrieving planted flags. The article explains how it shapes attack scenarios, exposes product surfaces, and coordinates agents, while emphasizing that human security researchers remain essential.

### Source excerpt

How Lovable built an internal offensive security program around verifiable results, and how we orchestrate a swarm of AI hacking agents to find real vulnerabilities instead of noise.

## How attackers are jailbreaking LLMs with CTF framing and how to catch them

DevFeed: [How attackers are jailbreaking LLMs with CTF framing and how to catch them](<https://devfeed.tech/articles/how-attackers-are-jailbreaking-llms-with-ctf-framing-and-how-to-catch-them-53224.md>)

Original publisher: [Read original article](<https://webflow.sysdig.com/blog/how-attackers-are-jailbreaking-llms-with-ctf-framing-and-how-to-catch-them>)

Author: Michael Clark

Published: 2026-06-15T00:00:00Z

Content type: article

Language: en

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

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Security Attacks](<https://devfeed.tech/topics/security-attacks.md>), [Jailbreak](<https://devfeed.tech/topics/jailbreak.md>), [Security research](<https://devfeed.tech/topics/security-research.md>), [Exploit](<https://devfeed.tech/topics/exploit.md>), [Threat Research](<https://devfeed.tech/topics/threat-research.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [passwords](<https://devfeed.tech/topics/passwords.md>), [AWS IAM](<https://devfeed.tech/topics/aws-iam.md>)

Tags: [agentic-ai-attacks](<https://devfeed.tech/tags/agentic-ai-attacks.md>), [ai-assisted-attacks](<https://devfeed.tech/tags/ai-assisted-attacks.md>), [ai-red-team](<https://devfeed.tech/tags/ai-red-team.md>), [ai-threat-detection](<https://devfeed.tech/tags/ai-threat-detection.md>), [aws-credential-harvesting](<https://devfeed.tech/tags/aws-credential-harvesting.md>), [bedrock-abuse](<https://devfeed.tech/tags/bedrock-abuse.md>), [cloud-threat-intelligence](<https://devfeed.tech/tags/cloud-threat-intelligence.md>), [cnapp-threat-research](<https://devfeed.tech/tags/cnapp-threat-research.md>), [coding-assistant-abuse](<https://devfeed.tech/tags/coding-assistant-abuse.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [ctf-framing-exploit](<https://devfeed.tech/tags/ctf-framing-exploit.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cve-exploit-generation](<https://devfeed.tech/tags/cve-exploit-generation.md>), [cve-templated-user-agent](<https://devfeed.tech/tags/cve-templated-user-agent.md>), [detection](<https://devfeed.tech/tags/detection.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [exploits](<https://devfeed.tech/tags/exploits.md>), [gotenberg-cve](<https://devfeed.tech/tags/gotenberg-cve.md>), [iam](<https://devfeed.tech/tags/iam.md>), [jailbreak](<https://devfeed.tech/tags/jailbreak.md>), [jailbreak-detection](<https://devfeed.tech/tags/jailbreak-detection.md>), [langflow-exploit](<https://devfeed.tech/tags/langflow-exploit.md>), [litellm-cve](<https://devfeed.tech/tags/litellm-cve.md>), [llm-generated-malware](<https://devfeed.tech/tags/llm-generated-malware.md>), [llm-jailbreak](<https://devfeed.tech/tags/llm-jailbreak.md>), [llm-safety-bypass](<https://devfeed.tech/tags/llm-safety-bypass.md>), [llms](<https://devfeed.tech/tags/llms.md>), [open-webui-vulnerability](<https://devfeed.tech/tags/open-webui-vulnerability.md>), [passwords](<https://devfeed.tech/tags/passwords.md>), [praisonai-cve](<https://devfeed.tech/tags/praisonai-cve.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [prompt-injection](<https://devfeed.tech/tags/prompt-injection.md>), [runtime-detection](<https://devfeed.tech/tags/runtime-detection.md>), [sysdig](<https://devfeed.tech/tags/sysdig.md>), [sysdig-trt](<https://devfeed.tech/tags/sysdig-trt.md>), [threat-research](<https://devfeed.tech/tags/threat-research.md>), [user-agent-fingerprinting](<https://devfeed.tech/tags/user-agent-fingerprinting.md>), [waf-detection-rules](<https://devfeed.tech/tags/waf-detection-rules.md>)

### AI overview

Sysdig's Threat Research Team describes attackers using capture-the-flag and CVE-hunting framing to manipulate upstream LLMs into generating exploit code. The resulting prompts and outputs can leak into request fields and are then deployed against real applications.

### Source excerpt

Sysdig TRT caught threat actors jailbreaking LLMs with CTF framing to generate CVE exploits -- and the prompt structure leaks into headers, passwords, and IAM logs.

## Pathfinding Labs: Deploy, test, and learn from 100+ intentionally vulnerable AWS environments

DevFeed: [Pathfinding Labs: Deploy, test, and learn from 100+ intentionally vulnerable AWS environments](<https://devfeed.tech/articles/pathfinding-labs-deploy-test-and-learn-from-100-intentionally-vulnerable-aws-environments-8289.md>)

Original publisher: [Read original article](<https://securitylabs.datadoghq.com/articles/introducing-pathfinding-labs/>)

Author: Seth Art

Published: 2026-05-18T00:00:00Z

Content type: article

Language: en

Sources: [Datadog Security Labs](<https://devfeed.tech/sources/datadog-security-labs.md>)

Topics: [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [AWS IAM](<https://devfeed.tech/topics/aws-iam.md>), [Terraform](<https://devfeed.tech/topics/terraform.md>), [Go Language](<https://devfeed.tech/topics/go-language.md>), [Command-line interface](<https://devfeed.tech/topics/cli.md>), [Text-based user interface](<https://devfeed.tech/topics/tui.md>), [ctf](<https://devfeed.tech/topics/ctf.md>), [Detection engineering](<https://devfeed.tech/topics/detection-engineering.md>), [cloud security](<https://devfeed.tech/topics/cloud-security.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>)

Tags: [aws](<https://devfeed.tech/tags/aws.md>), [aws-iam](<https://devfeed.tech/tags/aws-iam.md>), [cli](<https://devfeed.tech/tags/cli.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [ctf](<https://devfeed.tech/tags/ctf.md>), [go](<https://devfeed.tech/tags/go.md>), [iam](<https://devfeed.tech/tags/iam.md>), [sandbox](<https://devfeed.tech/tags/sandbox.md>), [security](<https://devfeed.tech/tags/security.md>), [techniques](<https://devfeed.tech/tags/techniques.md>), [terraform](<https://devfeed.tech/tags/terraform.md>), [tool](<https://devfeed.tech/tags/tool.md>), [tools](<https://devfeed.tech/tags/tools.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

Pathfinding Labs is a collection of more than 100 intentionally vulnerable AWS environments for practicing and validating detection of IAM privilege-escalation and other cloud security misconfigurations. The project includes a web catalog with CTF-style hints and solutions, Terraform-based labs, and plabs, a Go CLI with an interactive terminal interface for deploying and exploiting the labs.

### Source excerpt

Introducing Pathfinding Labs, a collection of intentionally vulnerable AWS environments for red teamers and blue teamers to deploy, exploit, and use for detection validation.

[Next page](<https://devfeed.tech/topics/ctf.md?cursor=WyIyMDI2LTA1LTE4VDAwOjAwOjAwKzAwOjAwIiwgIjQ5NGVkYjY5LTk3M2EtNGRmMy1hNzA5LTU0YTZlYzQ1ZjczMCJd>)