AGENT SKILL
benchmark-harness
Use when creating or revising CUDA benchmark runners and result artifacts.
- License 许可
- Apache-2.0
- 最近更新
- 2026-07-15
- Stars
- ★ 15
- Forks
- 1
01 前言 Front Matter
- name
- benchmark-harness
- description
- Use when creating or revising CUDA benchmark runners and result artifacts.
03 SKILL.md 正文
Benchmark Harness
Use this skill when creating or updating benchmark runners.
Workflow
- Define benchmark shapes, dtypes, and device tier.
- Add warmup and timed iterations.
- Ensure outputs go to
artifacts/benchmarks/. - Include an environment snapshot reference when the comparison matters.
- Preserve machine-readable output.
Minimum Metrics
- median latency
- p95 latency
- throughput when meaningful
04 同仓库更多 Skills
deploy-app
Deploy the application to staging or production environments.
★ 15
deploy
Deploy the application to environments.
★ 15
building-native-ui
Complete guide for building beautiful apps with Expo Router. Covers fundamentals, styling, components, navigation, animations, patterns, and native tabs.
★ 15
create-section
Use when creating a new section for the AI Workshop presentation. Guides through title slide setup, layout choices, and content gathering. Activates when user wants to add a new topic section to the slideshow.
★ 15
ship-checker
Run a full pre-ship review for this repo. Use when asked if a change is ready to deploy, publish, or ship.
★ 15
lint-check
Run repository linters and report failures.
★ 15