菜单
Reviews text for weasel words, vague language, and corporate jargon using Amazon's writing principles. Returns flagged issues and a polished version. Use when drafting Slack updates, PR descriptions, Jira tickets, or any professional communication that needs to be direct and concrete.
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。
基于 SOC 职业分类
Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.
A relentless interview to sharpen a plan or design.
Grill the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.
Captures high-value learnings from a skill execution and applies targeted improvements to the SKILL.md. Generic: works for any skill. Invoke right after running a skill when steps diverged from the documented flow, new techniques were discovered, or undocumented edge cases appeared. Synthesizes from conversation context automatically.
Audits and rewrites CLAUDE.md files. Use when the user asks to review, optimize, rewrite, standardize, or improve a CLAUDE.md file. Triggers include "audit claude.md", "otimiza o claude.md", "melhora o claude.md", "padroniza claude.md", "review CLAUDE.md". Also use when someone says "clean up my docs", "my claude context is messy", or asks to "initialize" context files in a project.
Pre-close session review. Scans for loose ends before ending a conversation — documentation gaps, unsaved memory, open checklist items, uncommitted git changes, and notification gaps. Reports findings grouped by category; never acts without confirmation. Invoke at the end of any work session before signing off.
| name | format-amznstyle |
| description | Reviews text for weasel words, vague language, and corporate jargon using Amazon's writing principles. Returns flagged issues and a polished version. Use when drafting Slack updates, PR descriptions, Jira tickets, or any professional communication that needs to be direct and concrete. |
Review the text for writing quality. Apply Amazon's writing standards: no weasel words, no vague quantifiers, no corporate jargon. Every claim must be specific and concrete.
Vague quantifiers without a number: significant, significantly, meaningful, many, most, some, better, improved, enhanced, optimized, increased, decreased (when no number follows)
Form-without-content adjectives: structured (describes format, not value: ask "what does it contain?"), complete (as in "complete picture": say what was missing before), stable (without a comparison baseline: say what was failing), automated (when the interesting part is what it does, not that it runs automatically), comprehensive, thorough, detailed (without specifics)
Corporate jargon: leverage, leveraged, synergy, synergies, robust, scalable, world-class, best-in-class, innovative, disruptive, seamless, cutting-edge, utilize, holistic, streamline, empower, enable, align, drive (as in "drive results")
Empty qualifiers: very, quite, rather, somewhat, fairly, pretty (as in "pretty good")
Filler outcome phrases: "complete picture", "single source of truth" (without saying what was fragmented before), "anyone on the team can use" (says nothing: who couldn't use it before?), "gives leadership visibility into" (say what they can now decide), "end-to-end" without specifying the two ends
Weasel phrases: going forward, at the end of the day, circle back, touch base, move the needle, at this point in time, in order to (replace with "to")
Weak attribution: we believe, we feel, we think, it seems, it appears (without data)
Passive voice that hides the actor: "was implemented by", "has been resolved", "is being addressed"
Issues found For each flagged item: quote the phrase, explain why it's weak, and suggest a concrete rewrite. If a number would fix it, say so explicitly.
Polished version The full text rewritten with all fixes applied. Keep the same structure and length.
