<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Keryn Headlines: SAST</title><link>https://keryn.dev/headlines/tag/sast/</link><description>Static analysis: finding flaws in source code before it runs.</description><language>en</language><lastBuildDate>Sun, 22 Nov 2026 15:38:32 GMT</lastBuildDate><atom:link href="https://keryn.dev/headlines/tag/sast/feed.xml" rel="self" type="application/rss+xml"/><item><title>AI SAST remediation: fixing vulnerabilities in first-party code</title><link>https://endorlabs.com/learn/ai-sast-remediation-fixing-vulnerabilities-in-first-party-code</link><guid isPermaLink="false">keryn-headlines-event-ev_acad89e2bfec</guid><pubDate>Fri, 20 Nov 2026 02:59:00 GMT</pubDate><category>Guide</category><category>Remediation</category><category>SAST</category><description>Learn how AI SAST remediation fixes vulnerabilities in first-party code with context-aware fixes, plus how to verify them and measure MTTR and fix rate.</description></item><item><title>Why the Harness Matters: Early Lessons From the Evo Code Scanning Agent Preview</title><link>https://snyk.io/blog/why-harness-matters-evo-code-scanning-agent</link><guid isPermaLink="false">keryn-headlines-event-ev_b7e22ed796aa</guid><pubDate>Thu, 19 Nov 2026 04:00:00 GMT</pubDate><category>Research</category><category>Agentic SDLC</category><category>SAST</category><description>In early tests, Evo Code Scanning Agent’s security harness found more known vulnerabilities than Claude Security and Codex Security at the same compute budget, with lower precision. See what the results reveal about the role of the harness in AI code scanning.</description></item><item><title>Reachability analysis for SAST findings: separating real paths from pattern matches</title><link>https://endorlabs.com/learn/reachability-analysis-for-sast-findings</link><guid isPermaLink="false">keryn-headlines-event-ev_02e0d9466410</guid><pubDate>Thu, 19 Nov 2026 03:22:08 GMT</pubDate><category>Guide</category><category>SAST</category><description>Reachability analysis for SAST checks whether untrusted input can reach a flagged sink along a feasible path. See how it works, how it differs from SCA, and how to use it in triage.</description></item><item><title>How to run AI SAST in CI/CD and on pull requests</title><link>https://endorlabs.com/learn/ai-sast-in-ci-cd-and-pull-requests</link><guid isPermaLink="false">keryn-headlines-event-ev_1752f72d3a07</guid><pubDate>Thu, 19 Nov 2026 03:02:42 GMT</pubDate><category>Guide</category><category>SAST</category><description>Run AI SAST where code gets reviewed. Scan placement, diff-aware PR scans, PR comments, merge policies, CI secrets, and metrics, with an example pipeline.</description></item><item><title>AI SAST best practices: 10 ways to run AI-powered static analysis well</title><link>https://endorlabs.com/learn/ai-sast-best-practices</link><guid isPermaLink="false">keryn-headlines-event-ev_32826973f897</guid><pubDate>Thu, 19 Nov 2026 02:23:30 GMT</pubDate><category>Guide</category><category>SAST</category><description>Ten practical AI SAST best practices: where to scan, how to trust findings, how to guard coding agents, and which metrics prove it&#39;s working.</description></item><item><title>Characterizing GitHub Security Actions Alerts For Web Application Tutorials</title><link>https://arxiv.org/abs/2611.23967</link><guid isPermaLink="false">keryn-headlines-event-ev_1b693251de52</guid><pubDate>Wed, 18 Nov 2026 07:05:33 GMT</pubDate><category>Research</category><category>SAST</category><description>The high frequency of cybersecurity incidents involving web applications highlights the importance of secure software development practices. Static Application Security Testing (SAST) tools help identify potential vulnerabilities in source code without executing the code.…</description></item><item><title>Detecting business logic flaws and IDOR with AI SAST</title><link>https://endorlabs.com/learn/detecting-business-logic-flaws-and-idor-with-ai-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_020b5101126d</guid><pubDate>Wed, 18 Nov 2026 05:37:57 GMT</pubDate><category>Guide</category><category>SAST</category><description>IDOR and broken access control look like valid code, so rule-based SAST misses them. See vulnerable and fixed examples and how AI SAST finds them.</description></item><item><title>The four places AI runs in Mend SAST</title><link>https://mend.io/blog/ai-in-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_c7aa4a3f7fb1</guid><pubDate>Tue, 17 Nov 2026 19:31:41 GMT</pubDate><category>Announcement</category><category>SAST</category><description>AI in SAST: Four jobs AI does in Mend SAST and what each gives developers.</description></item><item><title>Taint Analysis via Heap-Aware Propagation</title><link>https://arxiv.org/abs/2611.20453</link><guid isPermaLink="false">keryn-headlines-event-ev_472db264729a</guid><pubDate>Mon, 16 Nov 2026 03:23:21 GMT</pubDate><category>Research</category><category>SAST</category><description>Taint analysis for C depends on how precisely the analysis tracks tainted values as they flow through memory objects. Without field sensitivity, a tainted object field can spuriously taint an entire object, producing significant false alarms. In this paper, we present a…</description></item><item><title>Preserve the Vulnerability, Break the Alert: Testing the Robustness of CodeQL Taint Tracking with Natural Rewrites</title><link>https://arxiv.org/abs/2611.17123</link><guid isPermaLink="false">keryn-headlines-event-ev_ec7e8346b6bd</guid><pubDate>Thu, 12 Nov 2026 17:16:08 GMT</pubDate><category>Research</category><category>SAST</category><description>Static application security testing (SAST) tools such as CodeQL serve as security gates in continuous integration and code review, and a gate is also read in reverse. When a previously reported alert disappears after a change, developers and the tools around them often take the…</description></item><item><title>AI SAST vs traditional SAST: what changes when static analysis can reason</title><link>https://endorlabs.com/learn/ai-sast-vs-traditional-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_336000c019e4</guid><pubDate>Thu, 12 Nov 2026 13:09:58 GMT</pubDate><category>Analysis</category><category>Remediation</category><category>SAST</category><description>AI SAST vs traditional SAST, compared: how reasoning changes detection of authorization flaws, false positives, triage, fixes, speed, and cost</description></item><item><title>How AI SAST works: detection, triage, and remediation agents</title><link>https://endorlabs.com/learn/how-ai-sast-works</link><guid isPermaLink="false">keryn-headlines-event-ev_e2810977fc21</guid><pubDate>Thu, 12 Nov 2026 13:09:58 GMT</pubDate><category>Guide</category><category>SAST</category><description>See how AI SAST works step by step: what the detection, triage, and remediation agents do, the context each one uses, and how they hand off.</description></item><item><title>Your model, our analysis: BYOK for Endor Labs AI SAST</title><link>https://endorlabs.com/learn/byok-for-endor-labs</link><guid isPermaLink="false">keryn-headlines-event-ev_175e49f05b2e</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Announcement</category><category>SAST</category><description>Your model, our analysis: BYOK for Endor Labs AI SAST</description></item><item><title>AI SAST: Code Security for the Agentic SDLC</title><link>https://endorlabs.com/learn/ai-sast-code-security-for-the-agentic-sdlc</link><guid isPermaLink="false">keryn-headlines-event-ev_199e4588a9a5</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Announcement</category><category>SAST</category><description>How Endor Labs AI SAST combines program analysis, agentic reasoning, and deterministic engineering to find and fix the flaws that matter in AI-written code.</description></item><item><title>Why Endor Labs AI SAST for C finds what other tools miss</title><link>https://endorlabs.com/learn/why-endor-labs-ai-sast-for-c-finds-what-other-tools-miss</link><guid isPermaLink="false">keryn-headlines-event-ev_1c1208a912fb</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Announcement</category><category>SAST</category><description>Endor Labs AI SAST now fully supports C. Scans run in the IDE and on the PR, catching vulnerabilities before code ships, for teams writing C faster than release-gated review can keep up.</description></item><item><title>How AI SAST Traced Data Flows to Uncover Six OpenClaw Vulnerabilities</title><link>https://endorlabs.com/learn/how-ai-sast-traced-data-flows-to-uncover-six-openclaw-vulnerabilities</link><guid isPermaLink="false">keryn-headlines-event-ev_3e379c599fcf</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Research</category><category>SAST</category><description>We discovered six vulnerabilities in OpenClaw using Endor Labs’ AI SAST data flow analysis and validated working exploits.</description></item><item><title>What is AI SAST? A guide to AI-powered static application security testing</title><link>https://endorlabs.com/learn/what-is-ai-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_5b50006f86b8</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Guide</category><category>SAST</category><description>AI SAST combines program analysis, dataflow tracing, and LLM reasoning to find logic flaws rule-based SAST misses. Learn how it works and how to evaluate tools.</description></item><item><title>AI code review vs SAST: what each catches and how to use both</title><link>https://endorlabs.com/learn/ai-code-review-vs-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_5f5f2477acdc</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Guide</category><category>Agentic SDLC</category><category>SAST</category><description>AI code review vs SAST: how detection, bug coverage, determinism, false positives, and cost differ, and why combining them catches more real risk.</description></item><item><title>Endor Labs’ AI SAST Finds Zero Day Memory-Amplification DoS in Anthropic’s buffa library</title><link>https://endorlabs.com/learn/endor-labs-ai-sast-finds-zero-day-cve-2026-55407-buffa</link><guid isPermaLink="false">keryn-headlines-event-ev_72639a49ba11</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Research</category><category>SAST</category><description>We identified an unbounded-allocation data flow in buffa&#39;s protobuf decoder that I developed into a ~22x memory-amplification denial of service for CVE-2026-55407</description></item><item><title>Benchmarking Endor Labs AI SAST: 2.6x more real vulnerabilities found than frontier models</title><link>https://endorlabs.com/learn/ai-sast-benchmark-2x-more-real-vulnerabilities</link><guid isPermaLink="false">keryn-headlines-event-ev_75ac33efd7de</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Research</category><category>SAST</category><description>Endor Labs AI SAST found 192 real vulnerabilities and the most vulnerabilities no other tool found. Coverage is why. Here&#39;s what each tool actually caught.</description></item><item><title>How Endor Labs AI SAST Learns What Matters to Your Organization</title><link>https://endorlabs.com/learn/how-endor-labs-ai-sast-learns-what-matters-to-your-organization</link><guid isPermaLink="false">keryn-headlines-event-ev_841e3fc924d4</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Announcement</category><category>SAST</category><description>How Endor Labs AI SAST Learns What Matters to Your Organization</description></item><item><title>AI SAST Finding: Path Traversal in OpenClaw via LLM Guardrail Bypass</title><link>https://endorlabs.com/learn/ai-sast-finding-path-traversal-in-openclaw-via-llm-guardrail-bypass</link><guid isPermaLink="false">keryn-headlines-event-ev_97221ae05e3e</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Research</category><category>SAST</category><description>How Endor Labs&#39; AI SAST engine identified a path traversal vulnerability in OpenClaw&#39;s apply_patch tool tracked as (GHSA-r5fq-947m-xm57)</description></item><item><title>SAST for AI-Generated Code: What Static Analysis Catches and Misses</title><link>https://endorlabs.com/learn/sast-for-ai-generated-code-what-static-analysis-catches-and-misses</link><guid isPermaLink="false">keryn-headlines-event-ev_a9acb97c3e91</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Guide</category><category>SAST</category><description>Learn how sast for ai generated code catches logic flaws, cuts false positives, and exposes what traditional static analysis still misses</description></item><item><title>What is AI code review? How it works, what it checks, and where agentic review fits</title><link>https://endorlabs.com/learn/what-is-ai-code-review</link><guid isPermaLink="false">keryn-headlines-event-ev_c7037977af1b</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Guide</category><category>Agentic SDLC</category><category>SAST</category><description>AI code review uses AI to review pull requests for bugs, quality, and security. Learn how it works, how it differs from SAST and linters, and what agentic review adds.</description></item><item><title>9 Best SAST Tools in 2027: Accuracy, Speed, and Noise Compared</title><link>https://endorlabs.com/learn/best-sast-tools</link><guid isPermaLink="false">keryn-headlines-event-ev_d366f76de02c</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Roundup</category><category>SAST</category><description>Compare the best SAST tools of 2027 by accuracy, language coverage, and false-positive noise. See how the top static analysis tools rank.</description></item><item><title>Best AI SAST tools in 2027: 9 options ranked and compared</title><link>https://endorlabs.com/learn/best-ai-sast-tools</link><guid isPermaLink="false">keryn-headlines-event-ev_ebd23f97a6d1</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Roundup</category><category>SAST</category><description>Nine AI SAST tools ranked for 2027, from Endor Labs to Checkmarx, Semgrep, and more, compared on detection, triage, fixes, and the context their AI reasons over.</description></item><item><title>10 Snyk Alternatives for Engineering &amp; Security Teams (2027)</title><link>https://endorlabs.com/learn/snyk-alternatives</link><guid isPermaLink="false">keryn-headlines-event-ev_ed20d4803459</guid><pubDate>Wed, 11 Nov 2026 15:43:28 GMT</pubDate><category>Roundup</category><category>SAST</category><description>Compare 10 Snyk alternatives for SCA, SAST and container scanning, with reachability, AI capabilities and fit by team size, build system and compliance needs.</description></item><item><title>CyberBinGym: Can AI Agents Turn Binary Patches into Real Exploits?</title><link>https://arxiv.org/abs/2611.10723</link><guid isPermaLink="false">keryn-headlines-event-ev_8ea5373dc7fd</guid><pubDate>Sat, 07 Nov 2026 13:31:11 GMT</pubDate><category>Research</category><category>AI Security</category><category>Remediation</category><category>SAST</category><description>Security fixes usually reach users as compiled updates, but deployment delays leave unpatched systems exposed. During this window, comparing patched and unpatched binaries can reveal how to exploit a vulnerability, turning a defensive update into actionable information for…</description></item><item><title>I Don&#39;t Care How You Hide It, I Care What You Do: Runtime Detection of Malicious ML Models</title><link>https://arxiv.org/abs/2611.08745</link><guid isPermaLink="false">keryn-headlines-event-ev_b142b9f53ea4</guid><pubDate>Thu, 05 Nov 2026 21:52:21 GMT</pubDate><category>Research</category><category>SAST</category><description>Pre-trained machine learning (ML) models shared through public platforms can be weaponized to execute malicious payloads during loading or inference. Existing scanners prove inadequate for detecting malicious models as they rely on static analysis and function denylists,…</description></item><item><title>When File Ingestion Misleads Language Models: Measuring and Mitigating Divergences in Office Documents</title><link>https://arxiv.org/abs/2611.05806</link><guid isPermaLink="false">keryn-headlines-event-ev_d34552b02409</guid><pubDate>Wed, 04 Nov 2026 01:32:31 GMT</pubDate><category>Research</category><category>SAST</category><description>Office documents are now routinely uploaded to large language models (LLMs) for processing. However, an LLM does not always see what a human sees, because it typically reads the text that an extractor produces from the document&#39;s underlying Office Open XML (OOXML) files.…</description></item><item><title>Towards Effective Fuzzing of EDK2 Components</title><link>https://arxiv.org/abs/2611.03854</link><guid isPermaLink="false">keryn-headlines-event-ev_a81cf3af025d</guid><pubDate>Mon, 02 Nov 2026 19:23:21 GMT</pubDate><category>Research</category><category>DAST</category><category>SAST</category><description>Securing bootloaders is critical to establishing the trusted computing base of modern systems. TianoCore&#39;s edk2, the de facto UEFI-based bootloader, is widely deployed but has recently been shown to contain numerous high-severity vulnerabilities. Existing approaches either…</description></item><item><title>Fixed Detectors, Moving Code: How Well Do Static Security Analyzers Hold Up over a Decade of C/C++ Evolution?</title><link>https://arxiv.org/abs/2611.03742</link><guid isPermaLink="false">keryn-headlines-event-ev_6f583988e63b</guid><pubDate>Mon, 02 Nov 2026 02:16:44 GMT</pubDate><category>Research</category><category>SAST</category><category>SCA</category><description>Static application security testing (SAST) tools are often deployed on the assumption that detectors configured today will remain effective on future code, including previously unknown vulnerabilities. We test this assumption using 20,000 vulnerable and fixed C/C++ cases from…</description></item><item><title>Structured Security Explanations for Learning-Based Vulnerability Detection</title><link>https://arxiv.org/abs/2610.37797</link><guid isPermaLink="false">keryn-headlines-event-ev_40e1993f4e7d</guid><pubDate>Tue, 27 Oct 2026 18:16:40 GMT</pubDate><category>Research</category><category>SAST</category><description>Software vulnerability detection remains difficult because existing approaches rely on predefined analysis rules, learn vulnerability patterns implicitly from code and labels, or use large language models (LLMs) to make the final security judgment. It remains unclear whether…</description></item><item><title>ElectroNG is now open source!</title><link>https://blog.doyensec.com/2026/10/27/electrong-oss.html</link><guid isPermaLink="false">keryn-headlines-event-ev_f3e19779c5b9</guid><pubDate>Mon, 26 Oct 2026 23:00:00 GMT</pubDate><category>Release</category><category>SAST</category><description>ElectroNG is now free and open source! Today we are releasing both the desktop application and its core analysis engine under the Apache License 2.0, opening up years of Doyensec’s work on Electron application security to the community. Why we are open-sourcing ElectroNG We…</description></item><item><title>SymLift: Neural Semantic Lifting for Compile-Free Vulnerability Detection in Context-Incomplete Code Fragments</title><link>https://arxiv.org/abs/2610.34431</link><guid isPermaLink="false">keryn-headlines-event-ev_64049bfd176b</guid><pubDate>Sun, 25 Oct 2026 16:38:15 GMT</pubDate><category>Research</category><category>SAST</category><description>Missing project context can prevent build-dependent analyzers from constructing program relations required for vulnerability detection, while neural predictions alone do not establish auditable derivations. We present SymLift, a compile-free framework centered on…</description></item><item><title>Securing Agent-Generated Code with Its Own Development History</title><link>https://arxiv.org/abs/2610.34098</link><guid isPermaLink="false">keryn-headlines-event-ev_49fb4e999e5f</guid><pubDate>Sun, 25 Oct 2026 06:51:47 GMT</pubDate><category>Research</category><category>Agentic SDLC</category><category>Remediation</category><category>SAST</category><description>Pull requests written by coding agents can pass a project&#39;s test suite while omitting a guard or a validation step. To find the missing condition, a reviewer currently relies on static analysis of the final code. An agent-generated change, however, is accompanied by a…</description></item><item><title>Enhancing LLM-Based Vulnerability Detection via Representation-Aware Logical Ensemble</title><link>https://arxiv.org/abs/2610.33756</link><guid isPermaLink="false">keryn-headlines-event-ev_72c2999ccd74</guid><pubDate>Sat, 24 Oct 2026 15:55:52 GMT</pubDate><category>Research</category><category>SAST</category><description>Automated vulnerability detection in C source code is a long-standing challenge, as traditional static analysis tools often trade precision for scalability and struggle with complex vulnerability patterns. Recently, large language models (LLMs) have shown promising results;…</description></item><item><title>VulDec: Vulnerability Detection in Binaries via Neural Decompilation</title><link>https://arxiv.org/abs/2610.33716</link><guid isPermaLink="false">keryn-headlines-event-ev_3de9f48d24cd</guid><pubDate>Sat, 24 Oct 2026 15:02:51 GMT</pubDate><category>Research</category><category>SAST</category><description>Automated vulnerability detection in binary executables is challenging due to the lack of source code and the limited semantic information available at the assembly level. Despite significant progress in machine-learning-based code analysis, the field still lacks methodologies…</description></item><item><title>From Instructions to Exploits: Applying Program Analysis to Malicious Agent Skills Across Natural Language and Code</title><link>https://arxiv.org/abs/2610.33489</link><guid isPermaLink="false">keryn-headlines-event-ev_bc314a518a64</guid><pubDate>Sat, 24 Oct 2026 06:35:43 GMT</pubDate><category>Research</category><category>SAST</category><description>LLM-powered agents increasingly rely on third-party skills that combine natural-language instructions with executable code, creating supply-chain risks that may span both artifacts. Existing skill scanners often rely on rule matching or direct LLM judgments and provide limited…</description></item><item><title>Trusting the Tool: Asymmetric Deference of LLM Vulnerability Auditors to Static-Analysis Evidence</title><link>https://arxiv.org/abs/2610.32876</link><guid isPermaLink="false">keryn-headlines-event-ev_75b32128f776</guid><pubDate>Fri, 23 Oct 2026 16:45:10 GMT</pubDate><category>Research</category><category>Remediation</category><category>SAST</category><description>Pipelines that combine large language models (LLMs) with static analysis let the model consume the analyser&#39;s output as evidence, but whether the model distinguishes true from false evidence, and whether its scrutiny depends on what the evidence claims, has not been measured.…</description></item><item><title>SEMPRE: Semantic-Aware Protocol Reverse Engineering via Consistency-Driven Joint Optimization</title><link>https://arxiv.org/abs/2610.31844</link><guid isPermaLink="false">keryn-headlines-event-ev_154338f16c3b</guid><pubDate>Thu, 22 Oct 2026 19:27:55 GMT</pubDate><category>Research</category><category>SAST</category><description>Protocol reverse engineering (PRE) is essential for understanding undocumented or proprietary communication formats, enabling security-critical tasks such as vulnerability discovery, malware C2 analysis, and IDS signature generation. However, existing PRE techniques rely on…</description></item><item><title>JANUS: Agentic Cross-Language Fuzzing Harness Generation for Android Native Libraries</title><link>https://arxiv.org/abs/2610.30218</link><guid isPermaLink="false">keryn-headlines-event-ev_26f43d5e99c3</guid><pubDate>Wed, 21 Oct 2026 16:04:46 GMT</pubDate><category>Research</category><category>DAST</category><category>SAST</category><description>Modern Android apps extensively rely on native libraries, bridging Java/Kotlin and C/C++ through the Java Native Interface (JNI). This cross-language interoperation, however, exposes the memory-safe Android Runtime to the unsafe C/C++ world, creating an attack surface that has…</description></item><item><title>How Much Does the Harness Matter, and for Whom? A Multi-Model Study of Context in LLM-Driven Vulnerability Discovery</title><link>https://arxiv.org/abs/2610.28802</link><guid isPermaLink="false">keryn-headlines-event-ev_deb00cd543d2</guid><pubDate>Tue, 20 Oct 2026 18:25:47 GMT</pubDate><category>Research</category><category>Agentic SDLC</category><category>SAST</category><description>How much does an agent harness affect vulnerability discovery, and does its value depend on the model? We study 24 model-runner configurations using an open-source harness on a production codebase whose vulnerabilities were non-public at scan time. The pipeline combines static…</description></item><item><title>Self-Evolving Static Analysis via LLM Pseudo-Labeling</title><link>https://arxiv.org/abs/2610.30184</link><guid isPermaLink="false">keryn-headlines-event-ev_9c24e9ae7211</guid><pubDate>Tue, 20 Oct 2026 10:17:26 GMT</pubDate><category>Research</category><category>SAST</category><description>Security bugs often slip past static analysis not because they are impossible to detect, but because detection rules have not yet been written for the code in front of us. New frameworks, APIs, and coding idioms continually create such gaps, leaving developers to discover…</description></item><item><title>Raise or Check? Predicting the Necessity of Additional Fuzzing for Continuous Fuzzing</title><link>https://arxiv.org/abs/2610.27828</link><guid isPermaLink="false">keryn-headlines-event-ev_0d35b7a775b2</guid><pubDate>Tue, 20 Oct 2026 08:18:57 GMT</pubDate><category>Research</category><category>DAST</category><category>SAST</category><description>Continuous fuzzing is used for vulnerability detection in Open Source Software (OSS). Large-scale services such as Google&#39;s OSS-Fuzz fuzz over 1,300 projects and have revealed more than 10,000 vulnerabilities. Despite this, a considerable time lag often exists between when a…</description></item><item><title>A Hybrid Information Flow Analysis Framework for Android Hybrid Apps</title><link>https://arxiv.org/abs/2610.27288</link><guid isPermaLink="false">keryn-headlines-event-ev_55f41da2dba2</guid><pubDate>Mon, 19 Oct 2026 21:03:49 GMT</pubDate><category>Research</category><category>SAST</category><description>The WebView interface in Android (hybrid) applications enables the native side (e.g., Java) to exchange data with the Web (e.g., JavaScript). This creates cross-language information flow channels that existing single-language analyzers cannot analyze. Existing hybrid app…</description></item><item><title>How Effectively Do Security Skills Guide Coding Agents in Vulnerability Detection in Linux Kernel?</title><link>https://arxiv.org/abs/2610.25705</link><guid isPermaLink="false">keryn-headlines-event-ev_4800eb9d693c</guid><pubDate>Mon, 19 Oct 2026 02:50:45 GMT</pubDate><category>Research</category><category>Agentic SDLC</category><category>SAST</category><description>Large language model (LLM) coding agents, such as Claude Code, are increasingly used for security auditing, with task-specific skills providing reusable vulnerability-analysis procedures. However, their effectiveness in detecting real vulnerabilities and the reasons for their…</description></item><item><title>Combining LLM and Dynamic Analysis for Universal Testing of Static Application Security Testing Tools</title><link>https://arxiv.org/abs/2610.21309</link><guid isPermaLink="false">keryn-headlines-event-ev_10a344f6887f</guid><pubDate>Thu, 15 Oct 2026 05:57:29 GMT</pubDate><category>Research</category><category>SAST</category><description>Static application security testing (SAST) tools help detect potential software vulnerabilities, but they themselves contain faults undermining reliability and usability. Automated testing can address these issues, yet current techniques suffer from limited program…</description></item><item><title>FreeSBee: A Constant-Time Compilation Pipeline</title><link>https://arxiv.org/abs/2610.20171</link><guid isPermaLink="false">keryn-headlines-event-ev_fab2030e7f1a</guid><pubDate>Wed, 14 Oct 2026 13:20:04 GMT</pubDate><category>Research</category><category>SAST</category><description>Constant-time execution is an important security property, especially for IoT devices, that prevents the leakage of sensitive information over timing side-channels. However, achieving constant-time execution is very hard when going down to the cycle level, at least on realistic…</description></item><item><title>Mission (Im)Possible: Towards Autonomous Discovery of Autopilot Logic Flaws in Open-Source Drone Frameworks</title><link>https://arxiv.org/abs/2610.19566</link><guid isPermaLink="false">keryn-headlines-event-ev_01bb544b128f</guid><pubDate>Wed, 14 Oct 2026 04:30:11 GMT</pubDate><category>Research</category><category>SAST</category><description>Autonomous drones are increasingly deployed in safety-critical tasks, yet their autopilot software can still contain subtle logic flaws that lead to unsafe behaviors. These flaws are difficult to discover with existing techniques because they often depend on semantic…</description></item><item><title>Beyond Maliciousness: Static and Runtime Attack-Surface Profiling of Agent Skills</title><link>https://arxiv.org/abs/2610.19264</link><guid isPermaLink="false">keryn-headlines-event-ev_e4e5f99cf564</guid><pubDate>Tue, 13 Oct 2026 22:34:07 GMT</pubDate><category>Research</category><category>SAST</category><category>Trust</category><description>Agent skills are Markdown documents that a tool-using agent reads as instructions and then acts on with the privileges of the session that loaded them. Installing one grants authority (the standing to direct the agent, not new privilege), and the size of that grant is what we…</description></item><item><title>LLM-based Code Auditing by Inferring and Evaluating Developer Assumptions</title><link>https://arxiv.org/abs/2610.15729</link><guid isPermaLink="false">keryn-headlines-event-ev_a980a7cec0a1</guid><pubDate>Sun, 11 Oct 2026 16:07:53 GMT</pubDate><category>Research</category><category>SAST</category><description>When we see a null pointer check, we know that the developer believes that the pointer may not be initialized. If the same pointer is later used without a null pointer check, we may consider this a violation of the developer&#39;s belief and ultimately a bug. We develop such a…</description></item><item><title>SecReason: Vulnerability Detection and In-Stream Self-Repair for LLM Code Generation</title><link>https://arxiv.org/abs/2610.14453</link><guid isPermaLink="false">keryn-headlines-event-ev_18e3249e24b5</guid><pubDate>Sat, 10 Oct 2026 01:27:55 GMT</pubDate><category>Research</category><category>SAST</category><description>Large language models (LLMs) may introduce security vulnerabilities into generated code, which can persist into production when verification focuses on functional correctness. Existing defenses act before, during, or after generation, but none inspects the code during…</description></item><item><title>Frontier models found the vulnerabilities. Only the attacker found the chains.</title><link>https://snyk.io/blog/frontier-models-vulnerabilities-attacker-chains</link><guid isPermaLink="false">keryn-headlines-event-ev_b87055221917</guid><pubDate>Wed, 07 Oct 2026 04:00:00 GMT</pubDate><category>Analysis</category><category>Agentic SDLC</category><category>SAST</category><description>Static analysis found the flaws, but only live attack testing proved how they could be chained into breaches. A comparison of Evo COS, Claude Security, and Claude Code Security.</description></item><item><title>Introducing Wiz AI SAST: Application Security that Understands Your Code and Your Infrastructure</title><link>https://wiz.io/blog/introducing-wiz-ai-sast</link><guid isPermaLink="false">keryn-headlines-event-ev_1c4f0376563d</guid><pubDate>Tue, 06 Oct 2026 13:47:28 GMT</pubDate><category>Announcement</category><category>SAST</category><description>Find, validate, and fix complex business logic flaws with an AI code scanner backed by Wiz Research, operationalized within your existing security program.</description></item><item><title>CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?</title><link>https://arxiv.org/abs/2610.07557</link><guid isPermaLink="false">keryn-headlines-event-ev_c74819f7d84f</guid><pubDate>Tue, 06 Oct 2026 00:42:45 GMT</pubDate><category>Research</category><category>Remediation</category><category>SAST</category><description>Static-analysis checker synthesis requires agents to interpret a defect specification, inspect a repository, implement analyzer-specific logic, and refine the checker through repeated compilation and analysis feedback. Existing coding-agent benchmarks focus on tasks such as…</description></item><item><title>SoK: From Finding to Deployment: Systematizing the OS Kernel Bug Lifecycle</title><link>https://arxiv.org/abs/2609.23218</link><guid isPermaLink="false">keryn-headlines-event-ev_432814a478a5</guid><pubDate>Sat, 19 Sep 2026 21:25:07 GMT</pubDate><category>Research</category><category>DAST</category><category>Remediation</category><category>SAST</category><description>Automated kernel bug discovery has advanced rapidly. Continuous fuzzing and static analysis systems, such as syzbot, now expose Linux kernel bugs at a scale that downstream processes struggle to absorb. Yet a crash report is only the beginning. Before a bug is eliminated, it…</description></item><item><title>Auditing in the age of (good enough) AI</title><link>https://blog.trailofbits.com/2026/09/18/auditing-in-the-age-of-good-enough-ai</link><guid isPermaLink="false">keryn-headlines-event-ev_da19e5188bda</guid><pubDate>Fri, 18 Sep 2026 11:00:00 GMT</pubDate><category>Analysis</category><category>Agentic SDLC</category><category>SAST</category><description>Security firms have published numerous blog posts describing how they pointed their agent harness at a codebase and found dozens of bugs ( we’re one of them ). However, these posts tend to focus on agentic code review, which is just one aspect of how we use AI in our security…</description></item><item><title>Conjunctive Poisoning in AI Supply-Chain Applications</title><link>https://arxiv.org/abs/2608.15913</link><guid isPermaLink="false">keryn-headlines-event-ev_07b829c6191e</guid><pubDate>Sun, 16 Aug 2026 20:02:12 GMT</pubDate><category>Research</category><category>SAST</category><description>Large Language and Vision-Language Models are increasingly deployed through inference pipelines that include prompt wrappers (e.g., templates and post-processing scripts) and configuration metadata (e.g., JSON/YAML files) that together shape model outputs. While model weights…</description></item><item><title>What Is Checkmarx Fusion?</title><link>https://checkmarx.com/blog/what-is-checkmarx-fusion</link><guid isPermaLink="false">keryn-headlines-event-ev_d0bc142c98c9</guid><pubDate>Wed, 05 Aug 2026 18:40:36 GMT</pubDate><category>Release</category><category>SAST</category><description>Now available in Early Access for SAST within Checkmarx One, Checkmarx Fusion adds a new detection layer that advances vulnerability-finding accuracy beyond what rules-based scanning or an AI model can achieve alone.</description></item><item><title>We Now Build Code Faster Than We Can Understand It: The New Risk of AI Speed</title><link>https://checkmarx.com/application-security-trends/the-new-risk-of-ai-speed</link><guid isPermaLink="false">keryn-headlines-event-ev_49c0b637efb2</guid><pubDate>Tue, 28 Jul 2026 12:48:16 GMT</pubDate><category>Analysis</category><category>SAST</category><description>AI is generating code faster than teams can review it. Learn how hybrid SAST and Checkmarx Fusion help detect AI-era software security risks.</description></item><item><title>How we use /goal to find bugs in Patch the Planet</title><link>https://blog.trailofbits.com/2026/07/28/how-we-use-goal-to-find-bugs-in-patch-the-planet</link><guid isPermaLink="false">keryn-headlines-event-ev_a9322bd2c984</guid><pubDate>Tue, 28 Jul 2026 11:00:00 GMT</pubDate><category>Guide</category><category>Agentic SDLC</category><category>Remediation</category><category>SAST</category><description>Codex’s /goal feature amplifies bug hunting, but getting good results requires the right prompt, the right scope, and the right number of outcomes per run. For Patch the Planet , our joint initiative with OpenAI to find and fix bugs in open-source software, we pointed Codex at…</description></item></channel></rss>