appsec
SAMA’s Cyber Security Framework treats asset inventory as a foundational control. Organizations are expected to maintain an accurate inventory of assets, track configuration items, and understand how systems relate to each other. The logic is simple: if you don’t know what you own, you can’t secure it.
Research
Another expedition at Snapsec led us to Zendesk as our next target. The platform aligned well with our testing criteria, as its scope was extensive, the application surface was large, and the security team had a strong reputation for engaging constructively with researchers. With minimal scope restrictions and a feature-rich
News
1. What Are Supply Chain Attacks? Imagine you’re building a house. Instead of breaking into your finished house, a thief sneaks into the factory that makes your bricks or nails and swaps them for defective (or booby-trapped) ones. When you build with those “trusted” materials, your house gets infected
Jira Work Management is a widely used collaboration tool designed to help teams track all activity considered work. Work might include running projects, managing approval processes, performing daily and periodic tasks, creating documents, and Managing issues. Our team recently found an Stored XSS issue in Jira Work management which allowed
appsec
SAMA sets strict regulatory expectations through frameworks like its Cyber Security Framework, requiring organizations to maintain full visibility of assets, continuously manage vulnerabilities, and enforce clear remediation timelines. The approach is structured, risk-driven, and heavily focused on accountability at every level. Organizations are expected to do a few key things
Most organizations conduct security reviews on a schedule. Quarterly audits. Annual penetration tests. Periodic vulnerability scans. On paper, this creates the appearance of a well-governed security program. In reality, the attack surface does not follow review cycles. New infrastructure is deployed daily. Subdomains are created during product launches. Cloud resources
Research
For years, penetration testing has been treated as a milestone rather than a capability. Organizations schedule a test, receive a report, remediate a subset of findings, and move on, often for another six or twelve months. On paper, this satisfies compliance requirements and gives leadership a sense of assurance. In
It stars as a normal day. The ops lead had meetings, the coffee was lukewarm, and the ticket queue kept growing. “We’ll take it next week,” someone said. Next week became next month. Then, the breach call came. That’s the story repeated across industries: a small operational delay,
Back in March 2020 shuffling our private invites stock to crash into a program worthy of our time and excitement. In a while, we stumbled upon a program by name of Lark Technologies. Larksuite is a collaborrative platform where users can collaborate on various tasks. This product comprised of various
Modernizing Your Application Security
In modern enterprises, asset growth outpaces visibility. Infrastructure expands across multiple cloud providers, DNS zones multiply, APIs are deployed rapidly, and acquisitions introduce entirely new technology stacks. While security teams invest heavily in vulnerability detection, many organizations still depend on static asset inventories to define their security scope. Manual inventories
Modern security programs don’t fail because of missing scanners. They fail because of fragmented visibility. Domains live in one tool.APIs in another.IP ranges in spreadsheets.Certificates in someone’s inbox.Repositories in DevOps.Cloud buckets in yet another dashboard. Security teams think they have visibility. In reality,
In modern enterprises, vulnerability detection is no longer the bottleneck. With tools like Qualys, Nessus, Burp Suite, and Trivy, organizations generate thousands of findings daily across infrastructure, applications, containers, and cloud workloads. To manage this volume, security teams define SLAs: * Critical → 7 days * High → 14 days * Medium → 30 days On
For years, penetration testing has been treated as the ultimate proof of security maturity. Schedule a test, receive a report, fix the findings, and repeat the cycle next quarter or next year. On paper, this model appears rigorous. In practice, it no longer aligns with how infrastructure is built, changed,
The rapid rise of agentic AI systems has introduced a new class of security risk that sits uncomfortably between traditional malware, automation tooling, and cloud control planes. Clawdbot has become a useful reference point in this discussion—not because it is uniquely insecure, but because it reflects the broader security
Most organizations believe they understand their own infrastructure. They know their production environments, their internal networks, their approved services, and their deployed applications. Architecture diagrams exist. Asset lists exist. Ownership is documented-at least internally. Attackers see none of that. They do not start with your org chart, your CMDB, or
Most organizations believe they understand their security posture because their internal controls are strong. Firewalls regulate traffic, EDR monitors endpoints, SIEM correlates logs, and IAM governs identities. Inside the perimeter, visibility is high and response is mature. Yet breaches increasingly begin before any of those tools see activity. The reason
As large language models move deeper into productivity workflows, a subtle but dangerous shift is taking place. AI systems are no longer confined to chat interfaces or passive assistance. They are becoming orchestrators of data, capable of reading, summarizing, writing, and triggering actions across tightly integrated enterprise services. That power
Modern organizations have made significant investments in external visibility. Internet-facing assets are continuously scanned, domains and subdomains are enumerated, exposed services are detected, and configuration changes are monitored in near real time. From a tooling perspective, most security teams have more data than ever before about what is exposed externally.
Large language models are no longer experimental tools. They sit inside customer support systems, developer workflows, autonomous agents, internal knowledge bases, and even financial and operational pipelines. In many organizations, LLM-powered systems now make decisions, trigger actions, and interact with other services automatically. That shift has quietly created a new
Why Security Teams Struggle to Separate Real Threats from Background Noise Security teams today are not short on visibility.They continuously discover open ports, enumerate subdomains, detect new services, and track constant changes across the external attack surface. Dashboards are busy. Alerts are persistent. Yet one question consistently blocks action:
In November 2025, a quiet but massive security failure unfolded inside the world’s most widely used container registry. Flare’s analysis revealed that 10,456 Docker Hub images uploaded in a single month contained exposed secrets — from production cloud keys to AI model tokens and CI/CD access credentials.