Zed founder on agent sandboxes: aggressive isolation means manual auth steps
zeeg · x · 2026-09-12
Zed editor founder Nathan Soller (zeeg) shared hands-on impressions of a coding agent: its sandbox is fairly aggressive — it can't initiate auth for a CLI tool since the sandbox network isn't reachable, forcing manual work, which he calls a minor but real inconvenience. He also notes everything so far ran on the "high" setting, and he'll try "max" next.
Related event: Founders test Meta Muse Spark: fast but needs heavy hand-holding(5 posts)→
More from coding & agent
- When Facts Change: How Should an MCP Memory Server Update? Three Approaches Compared — izgorodin · 2026-09-12
- ADSP Ep. 303: Mark Saroufim on Open vs Closed Models, AI GPU Kernels and Autoresearch — blelbach · 2026-09-12
- Google releases free 2-hour course on agent graph & loop engineering — goyalshaliniuk · 2026-09-12
- Gemini CLI nightly patches indirect prompt injection and hardens sandbox filesystem — gemini-cli-robot · 2026-09-12
- Repurpose an old low-VRAM GPU just for mmproj in llama.cpp — an order of magnitude faster — inthesearchof · 2026-09-12
- Ollama vs llama.cpp: same request counts 85k vs 164k tokens in OpenCode — RadianceTower · 2026-09-12