CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

benchmark-checklist

Vet a measured number before reporting or acting on it: name the limiter, measure the claimed work, repeat with variance, rule out the observer, label every claim. For any speedup, regression, or score you are about to report.

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exemplary instruction-only skill: tight, well-sequenced, with explicit failure paths and a final gate that ties the checklist together. The only improvement space is in actionability — a worked example (e.g., a sample labeled benchmark report or a concrete measurement command) would close the gap between specific guidance and fully executable guidance.

DimensionReasoningScore

Conciseness

The body is lean and imperative with no padding and no explanations of concepts Claude already knows; each section contributes domain-specific judgment (e.g., "A change that 'improves' a metric the limiter does not govern measured noise, not work"), so every token earns its place per anchor 5.

5 / 5

Actionability

Concrete, specific guidance throughout ('at least five, more in noisy environments', 'Report the median and the spread', 'persist the raw output as an artifact under `.omc/`', labels 'measured, inferred, or guess') with only minor gaps — no example measurement commands or a sample labeled report — fitting anchor 4's mostly-executable-with-minor-gaps rather than anchor 5's fully copy-paste-ready coverage.

4 / 5

Workflow Clarity

A clearly sequenced five-step checklist with explicit per-step failure handling ('If you cannot name a plausible limiter, say so and treat the number as unexplained'; 'If the effect is smaller than run-to-run variance, the honest verdict is "no measurable change"') and a Verdict gate ('Only after all five sections pass does the number enter a reply'), which is exactly anchor 5's explicit validation with error-recovery feedback loops.

5 / 5

Progressive Disclosure

A self-contained, single-purpose checklist of roughly 60 body lines with well-organized sections (When to use, When not to use, five numbered steps, Verdict), no content that belongs in a separate file, and no nested references — per the simple-skill guideline this merits anchor 5 on well-organized sections alone.

5 / 5

Total

19

/

20

Passed

Description

88%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: it names five concrete actions, states an explicit trigger condition, and occupies a distinct niche. Its main gaps are missing natural trigger synonyms ('benchmark', 'faster', 'latency', 'throughput', 'eval') and slight breadth from including 'score' alongside performance terms.

Suggestions

Add missing natural trigger terms such as 'benchmark', 'faster', 'latency', 'throughput', or 'eval score' to the when-clause.

Consider whether 'score' alone is too broad; scoping it (e.g., 'eval score') would reduce overlap with eval-analysis skills.

DimensionReasoningScore

Specificity

Lists five concrete imperative actions ('name the limiter, measure the claimed work, repeat with variance, rule out the observer, label every claim') that comprehensively cover the skill's vetting process, matching the multiple-specific-actions anchor rather than the minor-gaps anchor below it.

5 / 5

Completeness

Both what (the five named vetting actions) and when ('For any speedup, regression, or score you are about to report') are explicitly stated with concrete trigger phrases; the 'For any... you are about to report' clause is equivalent to an explicit 'Use when...' clause, so this fits anchor 5 rather than anchor 4's vague-when case.

5 / 5

Trigger Term Quality

'speedup, regression, or score' are natural user phrases, but common synonyms like 'benchmark', 'faster', 'latency', 'throughput', and 'eval' are missing — notably the word 'benchmark' from the skill's own name never appears. Good coverage with a few natural terms missing fits anchor 4, not the comprehensive synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

The niche of vetting measured numbers before reporting is distinct with specific triggers, but 'or score' reaches into the eval domain, creating minor overlap risk with eval-analysis skills — anchor 4 rather than anchor 5's minimal-conflict profile.

4 / 5

Total

18

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository

Yeachan-Heo/oh-my-claudecode

Path

skills/benchmark-checklist/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.