菜单
Use tmux to launch and manage long-running jobs that must survive beyond the bash tool timeout. Use this skill when the user mentions long-running commands, background tasks, training jobs, large builds, or any command expected to exceed the bash tool timeout.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Analyze UT (unit test) results for a torch-xpu-ops PR. Use when asked to check test results, analyze CI failures, or evaluate test coverage for a PR. Produces a structured report of new failures, failure relevance, and new test coverage.
Rules for automatically determining which disable_* labels to apply to a PR based on the file paths changed. Used by the auto-label workflow.
Extract GPU ISA from oneDNN ngen-JIT kernels. This is the ONLY codegen path that bypasses the standard SYCL/SPIR-V/IGC stack. oneDNN uses its own native code generator (ngen) that directly emits GPU ISA bytes — no SPIR-V, no IGC, no zebin ELF, no .debug_line. Use when extracting ASM from oneDNN kernels (gemm_kernel, gen_conv_kernel), matmul, linear, conv, or SDPA-graph ops dispatched via mkldnn.
Extract GPU ISA from AOT-compiled SYCL binaries using NEO driver runtime dump. Use when the kernel is a SYCL symbol (_ZTS...) from an AOT binary such as libtorch_xpu.so, SYCL-TLA FMHA, or any DPC++ binary compiled with -fsycl-targets=spir64_gen -Xs "-device <gpu>".
Extract GPU ISA from JIT-compiled SYCL kernels. The binary only contains SPIR-V; IGC compiles it to native ISA at first launch. The zebin only exists in memory at runtime, so IGC_ShaderDumpEnable=1 is required to capture it. Use when extracting ASM from SYCL JIT kernels, torch-xpu-ops built with TORCH_XPU_ARCH_LIST=none, or standalone DPC++ without AOT flags.
Extract GPU ISA from Triton kernels on XPU. Triton compiles through the same IGC backend as SYCL (Triton IR → SPIR-V → IGC → zebin). Extraction is identical to sycl-jit: IGC_ShaderDumpEnable=1 captures the zebin at runtime. Use when extracting ASM from torch.compile fusions (triton_per_fused, triton_poi, triton_red), standalone @triton.jit kernels, or Inductor-generated XPU kernels.
| name | tmux-long-tasks |
| description | Use tmux to launch and manage long-running jobs that must survive beyond the bash tool timeout. Use this skill when the user mentions long-running commands, background tasks, training jobs, large builds, or any command expected to exceed the bash tool timeout. |
| license | MIT |
| metadata | {"tmux":"long-running jobs","bash":"background tasks"} |
The bash tool kills all processes (including & and nohup) when it times out. Use tmux for any job that may exceed the timeout.
Do NOT use &, nohup, or shell backgrounding for long-running jobs.
Before launching, you need:
If any are missing, ask the user and stop.
tmux list-sessions 2>/dev/null || echo "NO_TMUX_SESSIONS"
If tmux is not installed, stop and report.
If the target session does not exist, create it:
tmux new-session -d -s <session>
tmux new-window -d -n <job_name> -t <session>
tmux send-keys -t <session>:<job_name> 'cd /work/dir || exit 1; CMD 2>&1 | tee run.log' Enter
Use cd <dir> || exit 1 (not &&) so failures are visible in pane output.
If a window with the same name already exists, append a suffix (e.g., _2).
sleep 2
tmux capture-pane -p -t <session>:<job_name> | tail -5
If the shell prompt reappeared or an error is visible, the job failed immediately. Report to user.
tmux capture-pane -p -t <session>:<job_name> | tail -20
For full history, use the log file:
tail -20 /work/dir/run.log
tmux send-keys -t <session>:<job_name> C-c
After launching, report:
Job launched in tmux.
Session: <session> | Window: <job_name>
Log: <log_file>
Check: tmux capture-pane -p -t <session>:<job_name> | tail -20
Stop: tmux send-keys -t <session>:<job_name> C-c