erinhavens.bsky.social @erinhavens.bsky.social · 12/09/2026you can now roll the dice on erinhavens.com and see my site in the style of Botticelli, Claude, and even a lil Jeff Koons if you're lucky. 🎲🎲 000
Reposted by @erinhavens.bsky.socialGitHub @github.com · 13/08/2026In Session 4 of the GitHub Secure Open Source Fund, 50 projects upgraded their security posture through AI-assisted workflows, maintainer expertise, GitHub security tools, and more. As AI changes how software is built, maintainers remain at the center of protecting the open source ecosystem. 👇github.blogWhat 50 open source projects taught us about security in the AI eraSee how the projects in the GitHub Secure Open Source Fund combined AI-assisted workflows, maintainer expertise, and more to improve project security. 32511
erinhavens.bsky.social @erinhavens.bsky.social · 26/08/2026So very happy to see this @github.com blog post from one of our applied sciences teams. Our intent was to share best practices, advice, and lessons learned along the way! Let us know if you'd like to see more content like this. ❤️ github.blog/ai-and-ml/ll...github.blogHow to evaluate LLMs before productionThese are the lessons we learned evaluating LLMs for real-world secret scanning. 000
erinhavens.bsky.social @erinhavens.bsky.social · 09/07/2026To help you understand ownership and impact of a leaked secret, @github.com extended metadata is now GA for secret scanning alerts. 🫡 @branon.dev for the UI changes across security alert types ✨ 211
Reposted by @erinhavens.bsky.socialCan @canyesilyurt.com · 02/07/2026Okay, GitHub just dropped secret scanning for public repos for enterprise users. Finally, a dedicated guard for those oopsie moments when secrets escape into the wild! This is huge for preventing accidental public exposure. 👀 #GitHub #InfoSec 031
erinhavens.bsky.social @erinhavens.bsky.social · 02/07/2026Secrets don’t respect boundaries… scanning shouldn’t either! Public monitoring attributes secrets back to enterprises, from arbitrary public repos, in real time. github.blog/changelog/20... 100
erinhavens.bsky.social @erinhavens.bsky.social · 13/06/2026AI-powered secret detection has become good at surfacing candidates. The harder question: when is a signal worth interrupting a developer? Better context, not more data. ✦ Learn more about our work at @github.com @microsoft.com to make secret scanning more precise: github.blog/security/mak...github.blogMaking secret scanning more trustworthy: Reducing false positives at scaleAlerts are more trustworthy and actionable when noise is reduced. See how we improved the verification step with context-aware LLM reasoning. 020
