doc-coauthoring

Guides collaborative document creation through context gathering, iterative drafting, and reader testing.

Updated Aug 5, 2026
One-click install
npx skills add https://github.com/amirbiron/claude-skills --skill doc-coauthoring-amirbiron
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doc-coauthoring
Source: https://github.com/amirbiron/claude-skills/tree/main/skills/doc-coauthoring
Command: npx skills add https://github.com/amirbiron/claude-skills --skill doc-coauthoring-amirbiron

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing documentation, proposals, and specs alone often produces documents full of blind spots, missing context, and sections that confuse readers. This Skill structures the co-authoring process so the final document actually works for its intended audience. ## Core Features & Use Cases - Three-Stage Workflow: Context Gathering (info dumps and clarifying questions), Refinement & Structure (section-by-section brainstorming, curation, and drafting), and Reader Testing (validating the doc with a fresh, context-free reviewer). - Iterative Section Drafting: For each section, generates 5-20 brainstormed options, curates them with the user, drafts the content, and refines it through surgical edits. - Blind-Spot Detection: Tests the finished document against predicted reader questions to catch ambiguity, false assumptions, and contradictions before publication. - Use Case: A product manager needs to write a decision doc for a migration proposal. The Skill gathers organizational context, drafts each section collaboratively, then stress-tests the doc with a fresh reviewer to ensure stakeholders will understand it. ## Quick Start Help me write a technical design doc for our new caching layer using the structured co-authoring workflow.

Frequently Asked Questions about doc-coauthoring

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a technical spec with Claude?

Start a structured co-authoring workflow that gathers your project context first, then drafts the spec section by section through brainstorming and curation. Each section is refined iteratively until the full document is reviewed for coherence and tested against reader questions.

How to test if my documentation is clear to readers?

Use reader testing: predict 5-10 questions real readers would ask, then have a fresh Claude instance with no prior context answer them using only the document. Wrong answers or confusion reveal blind spots to fix before publishing.

What types of documents does this workflow support?

It supports structured content such as technical specs, decision docs, PRDs, RFCs, proposals, and general documentation. The workflow adapts its section suggestions based on the document type and any template you provide.

Can I use this workflow without Claude Code sub-agents?

Yes. On claude.ai without sub-agent access, reader testing is done manually: you open a fresh conversation, paste the document, and ask the predicted reader questions yourself. The rest of the workflow works identically.

When should I skip the structured doc workflow?

Skip it for short, low-stakes writing where three stages of iteration add overhead without value. The workflow is designed for substantial documents where reader comprehension and completeness matter, and you can decline it to work freeform at any time.