CookiesForAI

doc-coauthoring

Doc Co-Authoring

Doc Co-Authoring is Anthropic's agent skill that takes you and an AI agent through a document in three stages: gather context, draft section by section, then test it on a fresh reader.

Agent skill by Anthropic Official · Updated

Install Doc Co-Authoring

In Claude Code, add Anthropic's skills marketplace

Terminal
/plugin marketplace add anthropics/skills

Then install the example-skills plugin, which holds this skill and Anthropic's other example skills

Terminal
/plugin install example-skills@anthropic-agent-skills

In Claude.ai, Anthropic says its example skills are already available on paid plans: switch them on under Customize > Skills.

Or paste this into your coding agent: Install the agent skill doc-coauthoring from github.com/anthropics/skills. A skill can include scripts that run on your computer, so read its source first.

What Doc Co-Authoring does

Doc Co-Authoring is an Anthropic skill for writing structured documents with an AI agent, such as technical specs, proposals, PRDs, design docs, decision docs and RFCs. It loads when you start a writing task like these. The agent first offers the workflow and explains its three stages: context gathering, refinement and structure, and reader testing. If you decline, it works freeform.

In the first Doc Co-Authoring stage the agent asks about the document type, audience, desired impact and any template, invites you to dump all your context, then asks 5 to 10 numbered questions about the gaps. It can pull context from connected tools such as Slack or Google Drive when they are available. In the second stage it builds the doc section by section: clarifying questions, 5 to 20 brainstormed options, your pick of what to keep, a draft, then targeted edits.

In Doc Co-Authoring's reader testing, the agent predicts 5 to 10 questions readers will ask and checks whether a fresh agent with only the document can answer them, also looking for ambiguity, hidden assumptions and contradictions. In Claude Code it runs this with sub-agents; in Claude.ai it gives you the steps to run it in a new chat. Sections with gaps go back to drafting.

When to use Doc Co-Authoring

  • You need to write a spec, proposal or decision doc and the context lives in your head, your notes or team threads.
  • You want to know whether a doc makes sense to someone who was not in the conversation.
  • You want to build a long document one section at a time instead of rewriting one big draft.

When to pick something else

  • You want an existing page checked for style and tone: use Writing Guidelines.
  • You need a weekly status update or a company newsletter: use Internal Comms.

Which agents Doc Co-Authoring works in

Anthropic documents Doc Co-Authoring for Claude Code, Claude.ai, Claude API.

Reader testing uses sub-agents where the agent has them, and the SKILL.md names Claude Code. Without sub-agents, as in Claude.ai, it gives you steps to run the test yourself in a fresh chat.

Doc Co-Authoring license

Doc Co-Authoring is published under No license stated. "No license stated" means exactly that: the doc-coauthoring folder has no LICENSE file, its SKILL.md has no license key, and the anthropics/skills repository has no root license (checked 2026-10-01). Without a license, copyright law reserves all rights to Anthropic. Anthropic distributes it in its example-skills plugin, but nothing grants you the right to copy, change or redistribute it.

  • Gmail (gws): gws-gmail is the Gmail agent skill in Google Workspace's gws CLI repository: it lets an agent send, read, reply to, forward and triage Gmail messages through the gws command line. (31,231 repository stars on October 2, 2026)
  • Writing Guidelines: Writing Guidelines is Vercel's agent skill for docs review: the agent checks your pages against Vercel's Writing Guidelines and lists each style or tone issue by file and line. (31,813 repository stars on October 2, 2026)
  • Internal Comms: Internal Comms is Anthropic's open-source agent skill for writing inside a company: the agent drafts 3P updates, company newsletters, FAQs and other internal messages in set formats. (179,326 repository stars on October 2, 2026)

More from Anthropic

  • Academy Guide: Academy Guide is Anthropic's agent skill that ends Claude's answers about using Claude with one or two matching Claude Academy courses or tutorials, and only when the match is strong. (179,326 repository stars on October 2, 2026)
  • Algorithmic Art: Algorithmic Art is Anthropic's open-source agent skill for generative art: the agent writes a short art philosophy, then builds an interactive p5.js piece with seeds and parameter sliders. (179,326 repository stars on October 2, 2026)
  • Brand Guidelines: Brand Guidelines is Anthropic's open-source agent skill that holds Anthropic's own brand colors and fonts, so an agent can give a document, slide deck or web artifact the Anthropic look. (179,326 repository stars on October 2, 2026)
  • Canvas Design: Canvas Design is Anthropic's open-source agent skill for static visual work: the agent writes a design philosophy, then makes a poster or art piece as a finished PNG or PDF file. (179,326 repository stars on October 2, 2026)

Lists that include doc-coauthoring