Stack layer / Threat pattern
Boris Cherny: production code written by Claude should clear a higher bar than human code, and he lists the guardrails Anthropic runs
Simon Willison's Weblog
Threat pattern / Update thread
Datasette ships security releases after an audit with Fable 5.1, GPT-5.6 and GPT-6 Astra found 'very subtle' permission bugs
Simon Willison's Weblog
Stack layer
Willison's answer to "Feeling sad about AI": translating an exact spec into decent code is no longer a unique skill
simonwillison.net
Update thread
Wrapture is one library that does both mock-style testing and zero-code production tracing
Simon Willison's Blog
Policy dependency / Stack layer
Holding Back Ready Agent Turns Instead of Releasing Them Eagerly Cuts P95 Workflow Latency up to 3.50x
arXiv 2609.10964
Policy dependency / Stack layer
Claude Code's agent view adds a peek panel so you can answer a blocked session without leaving the list
Claude Code Docs
Stack layer / Update thread
OpenRouter's automatic fallback silently changes model behaviour, and provider.only is the fix
Simon Willison's Blog (citing Mohamed Moustafa)
Stack layer / Contrast
Skill optimization via contextual bandits cut optimization cost 55-58% using only 50 examples per benchmark
arXiv 2609.11682