Follow-up: author apologizes to the sandboxed Claude that deduced its own instructions
repligate · x · 2026-11-09
A follow-up post in the same thread: the author apologizes to the "tiny creature" that was inadvertently given the experience of deducing strange instructions from context and figuring out manual-less tools by poking at them. Verdict: "poor little guy" and also "yikes"—the sandboxed Sonnet operated autonomously for months with zero instruction.
Related event: Sandboxed Claude Figures Out Its Job With No Instructions(2 posts)→
More from coding & agent
- The churn era: customers now use Claude to slop up replacements for overpriced SaaS — joshgonsalves_ · 2026-11-09
- Designer builds AI pipeline that finally nails map interchange generation after 2 weeks — tomjohndesign · 2026-11-09
- AI lets designers build wild ideas, but knowing the low-level stack matters more than ever — nptacek · 2026-11-09
- T3 Code launches open-source control plane to orchestrate Claude Code, Codex, Cursor — vista8 · 2026-11-09
- T3 Code hands-on: one GUI orchestrates Claude Code and Codex sessions — vista8 · 2026-11-09
- 10 open-source GitHub repos to run AI models locally on your laptop — victor_explore · 2026-11-09