DeepSeek Harness Plugin

xinming-chenniang/dsh-vram-watch

Category Development & Runtime Added 2026-11-22

Diagnostic VRAM sampler for DSH: samples GPU memory continuously and keeps the last hour in memory. After a CUDA out-of-memory or an unexplained slowdown it reports four things instead of a live gauge — baseline, peak and final dedicated VRAM; each process first → last → peak plus whether it is still running; how much spilled into shared system memory; and how far dedicated VRAM stayed above baseline after the task exited. Two tools: vram_report (analyse the in-memory window or a saved CSV, optionally write the window to disk) and vram_watch (status / start / stop). Windows with an NVIDIA GPU only: reads WDDM performance counters through PowerShell and nvidia-smi; no hooks, no service, no screen capture.

Install

# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)

dsh plugin --profile web add github:xinming-chenniang/dsh-vram-watch

Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).

Links

More in this category

View the whole category →

Community comments

Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.