PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address. KERYN HEADLINES Daily, Tuesday 5 May 2026 Retrospective Covers 4 May 2026 00:00 to 5 May 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-05-05/ TOP DEVELOPMENTS [1] Backdoored PyTorch Lightning package drops credential stealer BleepingComputer · Incident · https://bleepingcomputer.com/news/security/backdoored-pytorch-lightning-package-drops-credential-stealer [2] CanisterSprawl: pgserve Compromised on npm: Malicious Versions Harvest Credentials and Exfiltrate to a Decentralized ICP Canister StepSecurity · Incident · https://stepsecurity.io/blog/pgserve-compromised-on-npm-malicious-versions-harvest-credentials [3] elementary-data Compromised on PyPI and GHCR: Forged Release Pushed via GitHub Actions Script Injection StepSecurity · Incident · https://stepsecurity.io/blog/elementary-data-compromised-on-pypi-and-ghcr-forged-release-pushed-via-github-actions-script-injection [4] Progress warns of critical MOVEit Automation auth bypass flaw BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/moveit-automation-customers-warned-to-patch-critical-auth-bypass-flaw SCA [5] A Mini Shai-Hulud Has Appeared: Obfuscated Bun Runtime Payloads Hit SAP-Related npm Packages StepSecurity · Analysis · https://stepsecurity.io/blog/a-mini-shai-hulud-has-appeared [6] Announcing Dependabot Configuration Enhancements: Cooldown and Group Support StepSecurity · Analysis · https://stepsecurity.io/blog/announcing-dependabot-configuration-enhancements-cooldown-and-group-support DAST [7] FunFuzz: An LLM-Powered Evolutionary Fuzzing Framework arXiv · Research · https://arxiv.org/abs/2605.02789 AGENTIC SDLC [8] TRE Python binding — ReDoS robustness demo Simon Willison · Analysis · https://simonwillison.net/2026/May/4/tre-python-binding [9] Copirate 365 at DEF CON: Plundering in the Depths of Microsoft Copilot (CVE-2026-24299) Embrace The Red · Analysis · https://embracethered.com/blog/posts/2026/defcon-talk-copirate-365 [10] These Aren't the Reviews You're Looking For How Humans Review AI-Generated Pull Requests arXiv · Research · https://arxiv.org/abs/2605.02273 REMEDIATION [11] APIOT: Autonomous Vulnerability Management Across Bare-Metal Industrial OT Networks arXiv · Research · https://arxiv.org/abs/2605.02346 TRUST [12] When Agents Handle Secrets: A Survey of Confidential Computing for Agentic AI arXiv · Research · https://arxiv.org/abs/2605.03213 [13] Human-Provenance Verification should be Treated as Labor Infrastructure in AI-Saturated Markets arXiv · Research · https://arxiv.org/abs/2605.03210 [14] Phionyx: A Deterministic AI Runtime Architecture with Structured State Management and Pre-Response Governance arXiv · Research · https://arxiv.org/abs/2607.18246 MODELS [15] Self-Mined Hardness for Safety Fine-Tuning arXiv · Research · https://arxiv.org/abs/2605.03226 [16] Revisiting JBShield: Breaking and Rebuilding Representation-Level Jailbreak Defenses arXiv · Research · https://arxiv.org/abs/2605.03095 [17] MITRE ATLAS (releases) v5.6.0 MITRE · Release · https://github.com/mitre-atlas/atlas-data/releases/tag/v5.6.0 AI SECURITY [18] PIIGuard: Mitigating PII Harvesting under Adversarial Sanitization arXiv · Research · https://arxiv.org/abs/2605.03129 ADVISORIES IN THIS WINDOW 77 npm advisories · 66 critical · 2 high · 7 moderate · 2 low 26 Go advisories · 1 critical · 12 high · 11 moderate · 2 low 16 Packagist advisories · 1 critical · 8 high · 7 moderate 15 Maven advisories · 9 critical · 4 high · 2 moderate 13 PyPI advisories · 1 critical · 4 high · 6 moderate · 2 low 5 RubyGems advisories · 1 high · 3 moderate · 1 low 3 crates.io advisories · 2 critical · 1 low All advisories: https://keryn.dev/headlines/advisories/ Titles, links and short excerpts only; every item links to its source. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.