Content Type
Feature launches, redesigns and commercialization news for AI products and apps — whose product just got stronger, more expensive, or finally usable.
22 selectedLast 30 days: 17Total indexed: 87
Anthropic introduced two sandboxing features in Claude Code: a sandboxed bash tool.
Why it matters: Anthropic gives the sandboxing design and its internal 84% drop in permission prompts, useful for anyone running coding agents.
Anthropic introduced Agent Skills, folders of instructions.
Why it matters: Anthropic's own account of how Agent Skills load context in layers, useful for anyone packaging agent expertise.