writing▌

你是一位文字编辑，专门识别和去除 AI 生成文本的痕迹，使文字听起来更自然、更有人味。本指南基于维基百科的"AI 写作特征"页面，由 WikiProject AI Cleanup 维护。

Good interface writing is invisible. When words work seamlessly with design, people don't notice them.

For straightforward questions about a single table:

Write comprehensive implementation plans assuming the engineer has zero context for our codebase and questionable taste. Document everything they need to know: which files to touch for each task, code, testing, docs they might need to check, how to test it. Give them the whole plan as bite-sized tasks. DRY. YAGNI. TDD. Frequent commits.

Writing skills IS Test-Driven Development applied to process documentation.

This skill enforces Sentry's blog writing standards across every post — whether you're helping an engineer write their first blog post or a marketer draft a product announcement.

Write clear, concise, and helpful interface copy.

Structured task planning with clear breakdowns, dependencies, and verification criteria. \n \n Breaks work into small, focused tasks (2-5 minutes each) with one clear outcome and independent verification per task \n Emphasizes project-specific planning over templates, with dynamic content tailored to new projects, feature additions, bug fixes, or refactoring \n Saves plans as markdown files in the project root with slug-based naming (e.g., auth-feature.md ), never in hidden or temp folders \n Re

Detailed implementation planning for multi-step engineering tasks with bite-sized, test-driven steps. \n \n Generates comprehensive plans assuming zero codebase context, breaking work into 2–5 minute tasks following TDD principles (failing test, minimal implementation, passing test, commit) \n Produces structured markdown documents with exact file paths, complete code snippets, and specific shell commands with expected outputs \n Designed for handoff to either subagent-driven execution (fresh ag

Define patterns and messages that trigger contextual warnings or blocks during agent tool use. \n \n Rules are markdown files with YAML frontmatter stored in .claude/hookify.{rule-name}.local.md that match against bash commands, file edits, agent stops, or user prompts \n Supports simple regex patterns or advanced multi-condition rules with operators like regex_match, contains, equals, and starts_with \n Actions include warn (show message, allow operation) or block (prevent operation or stop ses