Content
75%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A highly actionable, well-structured reference: real commands for every flag, explicit semantics for each filter mode, and a compilation prerequisite. Its main weakness is token efficiency — the Examples section largely repeats the inline option examples — and the absence of any post-run validation guidance keeps workflow clarity one notch below the top.
Suggestions
Remove the standalone '## Examples' section (or cut it to only combinations not already shown, such as combining --run with --grep), since most of its commands duplicate the inline examples in the Options section.
Consolidate the repeated --suite examples ('--suite git', "--suite 'api*'") into the single Options-section code block instead of showing them in both places.
Add a brief verification step after running tests (e.g., how to interpret exit codes or re-run a failing suite with --grep) to close the workflow-clarity validation gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The "## Examples" section duplicates commands already shown in the Options section ("--run src/vs/workbench/services/search/test/browser/search.integrationTest.ts", "--grep \"TextSearchProvider\"", "--suite git", and "--suite 'api*'" each appear twice). There is no padding with concepts Claude already knows, but a whole redundant section means the body could be tightened by roughly a third, which is more than the minor trimming of the anchor-4 example. | 3 / 5 |
Actionability | Every option is paired with a copy-paste-ready executable command (e.g. "./scripts/test-integration.sh --runGlob \"**/search/**/*.integrationTest.js\""), edge behavior is explicit ("Extension host tests are skipped since these filters are node.js-specific"), and the examples cover the common cases including combined flags ("--run ... --grep \"should search\""). | 5 / 5 |
Workflow Clarity | Each invocation mode's behavior is unambiguously enumerated (no filters, "--run or --runGlob", "--grep alone", "--suite"), and a pre-flight checkpoint exists ("Ensure the `VS Code - Build` watch task is running or that compilation has completed"). It falls short of anchor 5 because there are no validation or failure-recovery steps after a run, and the body is too long for the simple-skill exception. | 4 / 5 |
Progressive Disclosure | A self-contained, well-sectioned SKILL.md with no nested references and no bundle files to mis-navigate; the content is compact enough to belong inline. Minor organization gaps remain: the duplicated Examples section and the long inline suite list could be consolidated or split out. | 4 / 5 |
Total | 16 / 20 Passed |