← All skills
profiling-analysis-computing
80 / 100 · Low risk
Fine to try.
By ascend-ai-coding/awesome-ascend-skills (GitHub) · data · 184 stars · scanned at commit 75ec89c
用于分析Ascend NPU计算性能瓶颈：高耗时算子筛选、指令占比透视、形状解析，以及对 kernel_details.csv 中 aic_mac_ratio≥80% 的算力瓶颈算子做 Roofline 理论/实际耗时核对并给出是否需优化与排查方向。当用户提到计算瓶颈、算子分析、aic_mac_ratio、理论耗时、Roofline、kernel_details 算力核对时使用。
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
The author does not mention this.
Found in 11 places.
Show 8 more
- skills/profiling/profiling-analysis/profiling-analysis-computing/SKILL.md, line 119
python scripts/kernel_roofline_compute_check.py --input-path "./p-perf-huawei-05_110439_20250728062428118_ascend_pt" --flops 313 --bandwidth 1.2 --bw-unit TB/s - skills/profiling/profiling-analysis/profiling-analysis-computing/SKILL.md, line 136
python scripts/extract_op_shapes.py --input "./p-perf-huawei-05_110439_20250728062428118_ascend_pt" --output "./output/matmul_shapes.csv" --op "matmul" - skills/profiling/profiling-analysis/profiling-analysis-computing/SKILL.md, line 141
python scripts/extract_op_shapes.py --input "./p-perf-huawei-05_110439_20250728062428118_ascend_pt" --output "./output/matmul_shapes.csv" --op "matmul" --html-f - skills/profiling/profiling-analysis/profiling-analysis-computing/SKILL.md, line 213
python scripts/extract_op_shapes.py --input <profiling数据目录> --output <输出目录>/<op_type>_shapes.csv --op <op_type> - skills/profiling/profiling-analysis/profiling-analysis-computing/scripts/op_perf_analysis_combine.py, line 65
result1 = subprocess.run(selector_cmd, capture_output=True, text=True) - skills/profiling/profiling-analysis/profiling-analysis-computing/scripts/op_perf_analysis_combine.py, line 97
result2 = subprocess.run(analyzer_cmd, capture_output=True, text=True) - skills/profiling/profiling-analysis/profiling-analysis-computing/scripts/op_perf_analysis_combine.py, line 127
result3 = subprocess.run(extract_cmd, capture_output=True, text=True) - skills/profiling/profiling-analysis/profiling-analysis-computing/scripts/op_perf_analysis_combine.py, line 153
result4 = subprocess.run(roofline_cmd, capture_output=True, text=True)
Downloads and installs other software
The author does not mention this.
Found in 1 place.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'profiling-analysis-computing' ] || [ -L ~/.agents/skills/'profiling-analysis-computing' ]; then echo 'A skill folder named profiling-analysis-computing already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/ascend-ai-coding/awesome-ascend-skills' "$d" && git -C "$d" checkout --quiet '75ec89ce9d401be8ac95980e1c3f42f8d4315057' -- 'skills/profiling/profiling-analysis/profiling-analysis-computing' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/profiling/profiling-analysis/profiling-analysis-computing' ~/.agents/skills/'profiling-analysis-computing' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'profiling-analysis-computing' ~/.claude/skills/'profiling-analysis-computing' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for ascend-ai-coding/awesome-ascend-skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/ascend-ai-coding/awesome-ascend-skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-08 | 75ec89c | 100 |