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/TMDb --skill lint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。
Implement features and fix bugs test-first, the Canon TDD way (test list → one test → make it pass → refactor → repeat). Use when implementing any new feature, endpoint, model, or method, fixing a bug, or executing a plan — write the test list and a failing test BEFORE production code.
Capture durable, project-specific learnings from the work just done into the knowledge/ base — gotchas, implementation quirks, things looked up or web-searched, live-API behaviours, and design decisions (as ADRs). Use before opening a PR (it runs automatically in /deliver), or any time you've learned something worth remembering. Records only non-obvious, reusable facts; skips anything already in the repo, CLAUDE.md, or git history.
Take a plan all the way to a ready-to-merge pull request — review the plan (scaled to risk), implement it test-first, code-review and fix, run the CI gate, open the PR, and watch it green. Use after you have an approved plan (from plan mode, or the Plan agent); pass `issue <n>` to deliver a specific GitHub issue you name, or `next` to take the top startable issue off the project board's Ready column — either way it re-verifies, claims and plans the issue first. Invoking it with a plan already in hand is itself plan approval — it then runs autonomously to a single hard stop: ready-to-merge.
正在显示 SKILL.md
| name | lint |
| description | Lint code with swiftlint and swiftformat |
Run make lint from the project root to check code style. It runs eight
checks, in order: six Python scripts under Scripts/ — lint-witnesses
(check-defaulted-witnesses.py, protocol conveniences that would witness their
own requirement), lint-fixtures (check-fixtures.py, JSON fixture hygiene),
lint-curation (check-docc-curation.py, public methods missing from their
DocC page), lint-prose (check-prose-call-forms.py, code samples calling a
method that does not exist), lint-readme-version
(check-readme-version.py, README's .package(from:) vs the newest
CHANGELOG.md release) and lint-run-list (run-script-tests.py — despite
the target's name, every suite under Scripts/tests/: the /triage-issues
run-list builder, the /deliver selection-prose anti-drift cases, the
deliver-runfile.py writer, and the workflow cache-key/change-gate cases) —
then swiftlint --strict ., then swiftformat --lint ..
Each script enforces something no single-file linter can see, and a failure in
any of them is not a formatting problem: /format will not fix it.
Run this directly — it is fast and low-output, so delegating to a subagent would cost more (subagent overhead) than it saves.
If you see superfluous_disable_command errors on files you did not just change,
it is usually a swiftlint version-drift artifact (the pin is swiftlint 0.63.2 /
swiftformat 0.61.1), not a real violation. Run /format to auto-fix fixable
violations.
