Stack · 2 sessions

wsl2 sessions with coding agents

2 sessions on wsl2 with a coding agent: the agent failed at least once in 2, and 0 finished without a human stepping in.

More filters →

Claude Code·greenfield·Other·library2d

Built a custom Linux 3.0.1-based OS (TychaiOS) with custom init, userland tools and GOP framebuffer support

Takeaway: Next time I'd nail down the QEMU/OVMF/WSL boot pipeline first before layering framebuffer and GRUB experiments on top of it — half the pain was Windows/WSL path translation and stale OVMF vars, not the actual kernel or userland code.

4h 32m2interventionsC linux kernel qemu0 cmt

Claude Code·greenfield·Other·library3d

Built a custom Linux 3.0.1 OS 'TychaiOS' with a custom init, GOP snake, and GRUB/UEFI boot

Takeaway: Next time I'd start the prompt by stating upfront that I'm on WSL2/openSUSE driving QEMU from Windows, since half the wasted turns came from the agent trying WSL bash/PowerShell bridging that kept mistranslating paths.

4h 32m1interventionC linux kernel qemu0 cmt

Where agents get stuck on wsl2

Build failed repeatedly: implicit function declarations (missing _GNU_SOURCE), then a missing static libc for the link step.

The human: Cancelled a diagnostic tool call after exhausting patience with repeated debugging loops; told the agent to just rename the old working bzImage back and move on.

Built a custom Linux 3.0.1-based OS (TychaiOS) with custom init, userland tools and GOP framebuffer support →

Kernel-side build broke: implicit function declarations from missing _GNU_SOURCE, then linker failure from missing static glibc

The human: Saw the agent stuck running background WSL/PowerShell diagnostic scripts in circles trying to find why bzImage was missing; told it to just rename the old bzImage back manually and stop

Built a custom Linux 3.0.1 OS 'TychaiOS' with a custom init, GOP snake, and GRUB/UEFI boot →