Guide a doc from raw context to reader-tested clarity.
把零散上下文写成读者真正看得懂的文档。
Best for long-lived documents that real readers have to act on — RFCs, design docs, architecture write-ups, technical specs. Built around a reader-test loop: draft, hand to a fresh reader, see if they can answer the document's own questions, revise. Anthropic-authored, so the format and review conventions match how Claude actually consumes context. Not for one-pagers or release notes — release-briefing handles those with much less ceremony.
面对那种「真有人要照着做事」的长寿命文档时它最合适——RFC、设计文档、架构说明、技术规格。围绕读者测试循环展开：起草、丢给新读者、看 ta 能不能回答文档里自己提出的关键问题、再修订。Anthropic 出品，格式和评审约定和 Claude 实际消费上下文的方式对得上。一页发布说明那种小东西别用它，release-briefing 处理那种事开销更小。
Use npx skills add to install this skill into the selected agent. Phase 0 commands are generated from source rules, not verified.
npx skills add https://github.com/anthropics/skills/tree/main/skills/doc-coauthoring -g -a codex -yDrop `-g` to install project-locally
Create, test, and improve skills in an eval loop.
Distill how a person thinks into a reusable skill.
Sync the docs after the code changes, not three weeks later.
| this skill · 当前Anthropic Doc Coauthoring | Anthropic Skill Creator | Nuwa Skill | Docs Release | |
|---|---|---|---|---|
| Rating · 评分 | 4.7 | 4.9 | 4.7 | 4.7 |
| Stars · 星标 | 163k | 163k | 28k | 9.3k |
| Risk · 风险 | Low risk · 低风险 | Medium risk · 中风险 | Medium risk · 中风险 | Low risk · 低风险 |
| Best for · 最适合 | Design docs | New skill creation | Distill how a person thinks into a reusable skill. | README updates |
| Not for · 不适合 | Small changelog edits | One-off task automation | Workflows that require stronger human review than this catalog entry documents. | Live deploy verification |