Benchmark

Benchmark runtime performance on hardware. If the current branch has commits ahead of upstream/main or uncommitted changes, compares against the fork point (merge-base). Otherwise benchmarks current state only. Use when the user asks to benchmark, measure performance, or compare latency.

49 stars 102 forksGit & Code Review

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add hw-native-sys/simpler --skill benchmark
Manual install
git clone --depth 1 https://github.com/hw-native-sys/simpler.git /tmp/simpler
mkdir -p ~/.claude/skills
cp -r /tmp/simpler/.claude/skills/benchmark ~/.claude/skills/benchmark

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Benchmark Workflow

Benchmark runtime performance on Ascend hardware. Automatically detects whether to run a single benchmark or a comparison.

## Modes

| Condition | Mode | What happens |
| --------- | ---- | ------------ |
| 0 commits ahead AND no uncommitted changes | **Single** | Benchmark current state and report its runtime-specific timing columns |
| >= 1 commits ahead OR uncommitted changes | **Compare** | Benchmark merge-base (worktree) AND current workspace, show comparison table |

## Input

Optional benchmark arguments forwarded to `tools/benchmark_rounds.sh`:

```text
/benchmark
/benchmark -d 4 -n 50
/benchmark -d 4 -d 6
/benchmark --serial-orch-sched
```

Extra arguments (`-n`, `-r`, `--serial-orch-sched`, etc.) are forwarded to
`tools/benchmark_rounds.sh`.

### Device arguments (`-d`)

The `-d` flag specifies NPU device IDs.

**Hard rule: one benchmark process per device at any time.** Never run two benchmark processes on the same `-d` device concurrently — not two runtimes, not baseline + current, nothing. This prevents resource contention and ensures stable measurements.

On a shared hardware host, the complete single or compare run must execute
inside one `task-submit` allocation. Let that allocation own every device for
the full sequence; do not submit one job per example or run the script bare.

| `-d` count | Compare mode behavior |
| ---------- | --------------------- |
Read the full skill on GitHub

Details

Skill name
benchmark
Path
.claude/skills/benchmark/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Benchmark skill do?

Benchmark runtime performance on hardware. If the current branch has commits ahead of upstream/main or uncommitted changes, compares against the fork point (merge-base). Otherwise benchmarks current state only. Use when the user asks to benchmark, measure performance, or compare latency.

How do I install the Benchmark skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add hw-native-sys/simpler --skill benchmark. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Benchmark popular?

It lives in hw-native-sys/simpler, which has 49 GitHub stars and 102 forks.

Related skills

Checkout PR

49

hw-native-sys/simpler

Add a remote for a given PR's head repository and checkout a local pr-xxx-work branch. Use when the user wants to work on someone else's PR locally.

Git & Code Review102

Clean Branches

49

hw-native-sys/simpler

Remove stale local and remote git branches that have been merged into main. Detects regular merges and squash merges. Cleans up fork remote branches and prunes tracking refs. Use when the user asks to clean up branches, remove merged branches, or tidy up git.

Git & Code Review102

Core Swimlane

49

hw-native-sys/simpler

Produce a Core swimlane (the L0/intra-core AICore pipeline) for one task — a single kernel or a mix — via the dump-driven `simpler_setup.tools.core_swimlane` tool. Use when the user asks to "run/produce a core swimlane", "trace a task's intra-core pipeline", profile why one AICore task is slow inside the core(s), or needs help choosing the tool's manual flags (`--func-id`, `--restore-arg`, `--set-arg`, `--spmd-block-num`, `--case`). The tool captures real per-task args from an args dump and auto-generates the `msprof op simulator` replay — no hand-authored workspace. For a hand-authored single

Productivity & Notes102

Github

44

iamclancyliang/nanobot-eino

Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.

Git & Code Review

Alterlab Redcap Cdisc

87

AlterLab-IEU/AlterLab-Academic-Skills

Designs validated research data-capture instruments and aligns them to CDISC submission standards. Builds REDCap projects from a requirements spec: instrument and field design, the 18-column data dictionary (Variable/Field Name, Form Name, Field Type, Choices, Branching Logic, Text Validation, Identifier?), field validation (date_ymd, integer, number, email, phone), branching/show-field logic, longitudinal events and survey settings, and lints a data dictionary for common errors. Maps a study to CDISC: CDASH collection fields, SDTM domain mapping (special-purpose DM; Events AE, MH; Findings VS

Git & Code ReviewIn 1 list18

Create Issue

49

hw-native-sys/simpler

Create a GitHub issue following the project's issue templates. Classifies the issue type, fills required fields per template, and creates it via gh CLI. Use when the user wants to file a bug, request a feature, or create any GitHub issue.

Git & Code Review102

Browse the most popular Claude skills, all git & code review skills, or write your own with the skills generator.