用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/adamayoung/popcorn --skill test-single命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | test-single |
| description | Run a specific test target or test class |
| arguments | $ARGUMENTS |
Run a subset of the app's test suite by specifying a test target, class, or method
($ARGUMENTS). <specifier> is the -only-testing value: a test target,
Target/ClassName, or Target/ClassName/methodName.
Preferred — Xcode MCP (xcode), when running inside Xcode. Call it directly:
tabIdentifier from mcp__xcode__XcodeListWindows.mcp__xcode__GetTestList (with the
tabIdentifier) to find targets/classes.mcp__xcode__RunSomeTests with the tabIdentifier and the specifier(s)
matching $ARGUMENTS. On a build failure, call mcp__xcode__GetBuildLog with the
tabIdentifier and severity: "error".Fallback — make, when the MCP isn't available. For a long-running run you'll block on (e.g. a /deliver pre-PR gate step), prefer a backgrounded Bash (run_in_background: true) that redirects to a log and appends ; echo "EXIT=$?", then — on the completion notification — grep the log for the EXIT= marker plus failing tests (targeted greps keep the verbose log out of context); it self-reports once on exit, whereas a Haiku subagent driving its own poll loop tends to return premature "still waiting" notes. Otherwise, delegate to a Haiku subagent
(subagent_type: general-purpose, model: haiku) so output stays out of your context.
Do not run it yourself. Prompt the subagent to:
Run `mkdir -p .build && make test TEST_TARGET=<specifier> > .build/last-test.log 2>&1`,
check the exit status, and report ONLY:
- Status: passed or failed
- Counts: total / passed / failed
- Each failing test as `SuiteName/testName` with its `file:line` and message (omit if none)
- On failure, the log path `.build/last-test.log`
Do not paste passing-test output or raw logs.
Run the command yourself, immediately, in the foreground — never reply before it
has finished, and never defer, poll, or say you are waiting.
You are report-only: apart from the log file the command writes, do not create,
edit, or delete any file, and never attempt to fix what you find — report it.
MovieDetailsFeatureTests # whole target
MovieDetailsFeatureTests/MovieDetailsViewModelTests # a class/suite
MovieDetailsFeatureTests/MovieDetailsViewModelTests/loadSuccess # one test