What it is
doc-coauthoring is a skill that guides Claude through a structured workflow for writing documentation, proposals, technical specs, and decision docs. It runs three stages: Context Gathering, Refinement & Structure, and Reader Testing. Claude acts as an active guide, asking clarifying questions, brainstorming section content, drafting, and making surgical edits. It then checks the doc with a fresh Claude instance that has no context.
Who it's for
- Users starting a substantial writing task such as a PRD, design doc, decision doc, RFC, proposal, or spec
- Authors who want to verify a document works for readers, including readers who paste it into Claude
- Teams who want to pull context from tools like Slack, Teams, Google Drive, or SharePoint into a document
Requirements
Requirements
- Claude environment that can apply the skill; the source mentions Claude.ai, the Claude app, and Claude Code
- Optional integrations (e.g., Slack, Teams, Google Drive, SharePoint, or other MCP servers) to pull in context directly
- Artifact access (
create_file) or a working directory in which to create a markdown file - Sub-agent access (e.g., Claude Code) for automated reader testing; otherwise the user tests manually in a fresh Claude conversation
Examples
Stage 1 meta-context questions
Prompt1. What type of document is this? (e.g., technical spec, decision doc, proposal)
2. Who's the primary audience?
3. What's the desired impact when someone reads this?
4. Is there a template or specific format to follow?
5. Any other constraints or context to know?Expected output: Claude opens the context-gathering stage with these questions. The user can answer in shorthand or dump information however works best.
Shorthand answers to clarifying questions
Prompt1: yes, 2: see #channel, 3: no because backwards compatExpected output: After the info dump, Claude generates 5-10 numbered clarifying questions. The user can answer tersely like this, link more docs, or keep dumping context.
Curating brainstormed section options
PromptKeep 1,4,7,9
Remove 3 (duplicates 1)
Remove 6 (audience already knows this)
Combine 11 and 12Expected output: For each section Claude brainstorms 5-20 numbered options. The user curates them in this style, with brief justifications that help Claude learn priorities.
Requesting surgical edits
PromptRemove the X bullet - already covered by Y
Make the third paragraph more conciseExpected output: The skill asks users to describe changes rather than edit the doc directly, so Claude learns their style. Claude applies the edits with str_replace.
Manual reader-testing questions
PromptWhat in this doc might be ambiguous or unclear to readers?
What knowledge or context does this doc assume readers already have?
Are there any internal contradictions or inconsistencies?Expected output: Without sub-agents, the user pastes the doc into a fresh Claude conversation and asks these checks, in addition to the predicted reader questions.
Pros & cons
Pros
- Pro:Gives a clear, staged process that covers context transfer, section-by-section drafting, and verification
- Pro:Reader Testing with a fresh Claude catches blind spots before real readers see the doc
- Pro:Works with or without integrations, artifacts, and sub-agents, with documented fallbacks for each
- Pro:Keeps the user in control: they can decline the workflow, skip stages, or switch to freeform writing
Cons
- Con:Process-heavy, with many question and iteration rounds, which may be slow for short or simple docs
- Con:Automated reader testing needs sub-agent access; otherwise the user must run the tests manually in a separate conversation
Images
Related
Skill 
Skill algorithmic-art
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Skill brand-guidelines
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
Skill canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Skill claude-api
|-
Skill