2026-08-19
Microsoft open-sources ThinkingBox to test whether agents finish the job
Microsoft published ThinkingBox on August 19, 2026, an open-source sandbox that grades an agent on the state it actually leaves behind in a backend rather than on what it claims to have done. Agents run in isolated, stateful tool environments with a simulated user that answers follow-up questions. The paired ThinkingBox-Bench covers 507 policy-conditioned tasks across retail and ecommerce, travel and hospitality, auto insurance, neobank internal IT, and consulting IT and HR.
2026-08-06
Agent Plugins 1.0.0 ships as a vendor-neutral standard
Agent Plugins 1.0.0 was published on August 6, 2026 — an open specification for packaging Agent Skills and MCP servers as portable plugins. A plugin is a directory with a plugin.json manifest and fixed locations for its components, so one package loads across any compatible client. Vercel initiated the proposal; AWS, Anysphere, GitHub, Microsoft, and OpenAI refined it collaboratively. It is supported at launch in ChatGPT, Codex, Cursor, GitHub Copilot, Kiro, and VS Code.
