FEATURED
06:18 UTC
MCP 2.0 goes stateless, making agent tools easier and safer to ship
new feature deep dive
medium
MCP 2.0’s stateless design makes agent tooling a lot simpler to build and much safer to run in production.
anthropic
06:19 UTC
CISA’s 2026 SBOM update now covers AI and SaaS and requires hashes
policy legal enterprise
high
SBOMs now need hashes and must cover AI/SaaS—wire this into builds, deploy gates, and vendor contracts before it’s mandated.
github
06:19 UTC
GitHub Copilot agents get faster long-session handling and tighter approvals
new feature deep dive
medium
Copilot’s agent tooling is maturing: faster long-session handling, clearer approvals, and fewer flaky SDK edges for day-to-day use.
openai
06:20 UTC
OpenAI slashes Luna/Terra pricing; SDK adds better backoff and provenance checks
pricing plan changes
high
Price cuts make Luna/Terra strong defaults—pair them with the new SDK backoff and provenance to save cost without adding risk.
openai
06:22 UTC
Codex growing pains: scale bugs, VS Code extension hiccups, and the limits of AI on tribal knowledge
release problems outages controversies
medium
Treat Codex as prototype-grade today: useful, but watch scale bugs, extension stability, and the gap around undocumented system knowledge.
langchain
06:22 UTC
Agentic coding grows up: prove it, cap it, then scale it
trend pattern
medium
If you can’t prove what the agent did or cap its spend, you don’t have a system — you have a liability.
amazon-web-services
06:23 UTC
Nscale buys Anyscale: what it means for Ray teams and multi-cloud neutrality
policy legal enterprise
high
If you rely on Anyscale for Ray, build a cloud-neutral Plan B now.
anthropic
06:23 UTC
Stop defaulting to frontier LLMs: vCodeX’s auto-routing play to cut token burn
workflow use case
medium
Put a model router in front of your dev tools and stop paying frontier prices for lookup-level prompts.