一键导入
writing-prose
Use when writing documentation, READMEs, explanations, blog posts, or any prose meant for humans to read
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。
菜单
Use when writing documentation, READMEs, explanations, blog posts, or any prose meant for humans to read
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
This skill should be used when the user asks to "rebase all open PRs", "update all PRs with latest main", "make sure every open PR is up to date with main", "sync all open PRs onto main", "rebase all my PRs", or wants to bring every open pull request current with its base branch by rebasing in parallel git worktrees, resolving conflicts with a team of agents, and force-pushing.
Groom and prioritize a project backlog — verify each candidate issue's claims against the current codebase, pick the most important N, rewrite them to a ready-to-work standard (Problem/Goals/Decisions/Outcomes/Verification), move them to Ready on the board, and close obsolete issues. Use when asked to "groom the backlog", "triage the backlog", "prioritize issues", "move issues to ready", or "what should we work on next".
This skill should be used when the user asks to "summarize a note", "summarize this article", "replace #summarize-later", "add a summary to a note", "generate takeaways", or mentions summarizing Obsidian notes. Replaces the #summarize-later tag with a high-level summary and actionable takeaways.
This skill should be used when the user asks to "update the changelog", "add changelog entries", "what changed since last release", or when updating CHANGELOG.md with user-facing changes. Follows Keep a Changelog format and filters out internal-only commits.
This skill should be used when the user asks to "improve a prompt", "make this prompt better", "optimize this prompt", "refine this system prompt", or wants to apply concept elevation to compress and clarify LLM instructions.
Use when writing, reviewing, or improving system prompts for AI tools, agents, chatbots, or LLM-powered products. Trigger phrases include "write a system prompt", "create a system prompt", "system prompt for X", "improve this prompt", "review my prompt", "prompt engineering".
| name | writing-prose |
| description | Use when writing documentation, READMEs, explanations, blog posts, or any prose meant for humans to read |
Write like you talk. Then cut.
Every sentence must pass this test: "Would I say this to a friend?" If not, rewrite it.
AI-assisted writing runs longer than the writer would have produced by hand: paragraph after paragraph of plausible-sounding text that says little. Length is itself a tell, and every extra word taxes the reader.
Cut to what you'd write if you had to say it out loud to a busy person. This is not terseness: keep every idea the reader needs, drop the words that carry none. If a cut loses meaning, keep the words. If it only loses padding, cut.
Test: Is this longer than I'd have written without help? If yes, cut.
Usefulness = Correctness × Novelty × Importance × Strength
These multiply. Zero in any dimension = zero value.
For each sentence ask:
NEVER use emdashes (—). Use commas, periods, colons, semicolons, or parentheses instead. Rewrite any sentence that seems to need an emdash.
| Prefer | Avoid |
|---|---|
| Short, common words | Jargon, fancy vocabulary |
| Active voice | Passive constructions |
| Concrete nouns | Abstract nominalizations |
| "Use" | "Utilize" |
| "Help" | "Facilitate" |
| "About" | "Regarding" |
Germanic words beat Latinate ones. "Begin" not "commence." "End" not "terminate."
These signal bad prose. Rewrite immediately:
Writing that sounds good is more likely to be right. Rewriting for flow forces better ideas. You can't consciously make ideas worse while making prose better.
If a sentence sounds clumsy, the idea is probably muddled. Fix the thinking, not just the words.
# Bad
It is imperative that users be made cognizant of the fact that...
# Good
Users need to know that...
# Bad
After careful consideration of the various factors involved in the
decision-making process, we have determined that the optimal course
of action would be...
# Good
We decided to...
# Bad
The data validation error handling mechanism implementation...
# Good
How we handle validation errors...
# Bad
It is generally believed that this approach is somewhat better.
# Good
This approach is better because [specific reason].
Qualifications are not weaknesses; they are precision tools:
Bad qualification hedges against criticism. Good qualification sharpens the claim.
From Italian "saltimbocca" (leaps into mouth). Ideas should leap into the reader's head. They barely notice the words that got them there.
Simple language lets readers focus on ideas, not parsing prose.
| Principle | Action |
|---|---|
| Conversational | Would I say this to a friend? |
| Simple | Ordinary words, short sentences |
| Bold | Strong claims, qualified precisely |
| Honest | Simple prose exposes weak ideas |
| Rhythmic | Sound matches logic |
| Cut | Delete everything unnecessary |