The Document Foundation says LibreOffice 26.8 drew 1,031,162 downloads from its official site in its first week, the highest first-week figure for any LibreOffice release.
ClickHouse acquired security data platform RunReveal, folding in-house the expertise behind a product already built on its own database.
HashiCorp argues HCP Terraform's policy, identity, and audit controls must govern AI agents authoring infrastructure changes, pointing to its new tfctl CLI as part of that model.
GitHub Copilot's automated code review can now submit approvals that count toward merge requirements, but the capability ships disabled by default and requires an administrator to opt in at the enterprise, organization, or repository level.
Google rolled out Pics, an AI image generator and editor built natively into Workspace, powered by its Nano Banana model and aimed at Canva- and Adobe Express-style design tasks.
JetBrains' Compose Multiplatform 1.12.0 adds an experimental MCP server to Compose Hot Reload, letting AI agents reload, screenshot, and click through running apps.
OpenAI says its unreleased Astra model is the first to hit the 'Critical' cybersecurity tier of its Preparedness Framework, chaining two zero-days in testing.
Manifold Security found eight flaws across seven AI coding agents that let a repository's git config silently execute code the moment it is opened.
The ASF's FY2026 annual report tallies 10,225 committers, 302 projects, 11 new Top-Level Project graduates, and 1,709 releases for the year ended April 30, 2026.
Figma replaced Projects with folders that nest up to 10 levels deep starting August 3, 2026, but users on its own forum immediately flagged missing previews, non-persistent sorting, and broken API endpoints.
A U.S. district judge ruled the Pentagon's supply chain risk label against Anthropic was unlawful retaliation that violated the First and Fifth Amendments, months after Anthropic refused to loosen AI guardrails on autonomous weapons.
Harper's 5.2 release isolates database commits onto dedicated threads and adds a record cache, while a company benchmark argues single-runtime architecture beats a Vercel-based stack on personalized reads.