Spec Kit Templates v0.2.1: Kimi Code & Extensions
Spec Kit 0.2.1 adds Kimi Code agent support, .extensionignore filtering, and Codex command registration for extensions.
github
Updates von GitHub und GitHub Copilot
GitHub's Outage Postmortem: What Broke and What They're Fixing
GitHub published a detailed postmortem of three major outages in early 2025. Cache TTL changes, 10x client traffic spikes, and failover bugs exposed architectural limits. Here's what broke and what they're fixing.
GitHub Copilot SDK: Embed Agentic Execution in Your Apps
GitHub released the Copilot SDK — the same execution engine behind Copilot CLI, now embeddable in your apps. Multi-step planning, tool invocation, and error recovery without building orchestration from scratch.
Spec Kit Templates v0.2.0: Multi-Catalog Support
Spec Kit 0.2.0 adds simultaneous multi-catalog support, Tabnine CLI integration, and fixes hook event wiring for agent workflows.
GitHub Agentic Workflows: Security Architecture Deep Dive
GitHub Agentic Workflows isolates agents in containers with zero access to secrets, buffers all writes for vetting, and logs every action. Here's how the three-layer security architecture makes unsupervised agents safe for CI/CD.
GitHub Security Lab's AI Framework Found 80+ Vulnerabilities
GitHub Security Lab's open-source AI framework found 80+ real vulnerabilities by teaching LLMs to understand threat models first. Here's how it works and what it found.
GitHub Copilot Code Review: 60M Reviews and the Agentic Shift
GitHub's Copilot code review hit 60M reviews by rebuilding with an agentic architecture. It now handles 1 in 5 GitHub code reviews, stays silent 29% of the time, and helped WEX ship 30% more code. Here's how the new system works.