Claude for writers
Write with Claude without losing the writer.
A source-grounded workflow library for fiction, copy, editing, research, and publication decisions. Built for writers who want better drafts, not generic AI prose.
Editorial north star
The useful answer is not "write it for me."
The useful answer is a repeatable system: context, source boundaries, draft passes, critique, verification, and final human judgment. Claude Writes turns that system into practical guides with primary sources attached.
Core guides
Start here
Workflow
The best Claude writing workflow treats the model as a structured collaborator: give it context, constrain the task, request analysis before prose, then verify and rewrite under human control.
Fiction craftFiction
Use Claude as a developmental reader and scene mechanic. Keep premise, taste, voice, and final prose under human control; ask Claude to expose contradictions, generate options, and test emotional causality.
Revision systemEditing
Claude should not perform every edit at once. Assign one editorial layer at a time, require a diagnosis before rewrites, and keep the final accept/reject decision with a human editor.
Commercial writingCopywriting
Copywriting with Claude works best when the source of truth is a claims inventory. The model can generate angles and variants, but every promise needs proof, permission, and compliance review.
llms.txt, llms-full.txt, JSON-LD, citation blocks, source cards, and quote anchors.
Anchor this statClaude feature claims cite Anthropic or Claude docs; copyright and marketing claims cite government sources.
Anchor this statThe freshness log tracks source reviews and future changes to volatile Claude product surfaces.
Anchor this statFree tools
Build the workbench
Writing Workbench
Choose the writing form, goal, audience, source boundaries, and voice constraints; the workbench turns them into a copy-paste Claude prompt with a built-in revision loop.
Free editing toolRevision Planner
The planner inspects a draft in the browser, reports simple revision signals, and recommends focused Claude prompts for developmental, line, copy, and proof passes.
Free style toolStyle Guide Builder
Fill in voice, POV, tense, banned language, examples, and editing boundaries; the builder produces a paste-ready guide for Claude chats, projects, or CLAUDE.md files.
Full library
Guides for the writing desk
A prompt library for Claude writing workflows: briefs, source packets, fiction scenes, editing passes, copy variants, and final verification.
Fact-checkingFact-Checking Claude Drafts: A Source-First WorkflowA verification workflow for Claude-assisted writing: source packets, claim tables, quotation checks, current facts, and publication gates.
RightsClaude, Copyright, Privacy, and Disclosure for WritersA writer-focused guide to AI-assisted authorship, copyright registration, data privacy, disclosure, and responsible Claude use.
SetupClaude Project Setup for WritersHow to configure a Claude writing project with instructions, source packets, voice guides, skills, artifacts, and update routines.
VoiceHow to Preserve Your Writing Voice When Using ClaudeA practical workflow for using Claude without generic AI prose: voice inventories, annotated examples, exclusions, edit boundaries, and final human line judgment.
ResourcesClaude Writing Sources and Further ReadingPrimary sources for Claude writing workflows: Anthropic docs, Claude Help Center articles, privacy pages, Copyright Office guidance, and FTC AI-claim enforcement.
UpdatesClaude Writes Freshness LogA freshness surface for Claude Writes: source review dates, article updates, and weekly extension points for Claude writing workflows.
Independent publication
Not affiliated with Anthropic.
Claude is a product of Anthropic. Claude Writes is an independent educational site about writing workflows, source verification, and responsible publication.
Primary sources
Official guidance on clarity, context, examples, XML tags, roles, and long-context prompt structure.
Claude Help CenterWhat are projects?March 16, 2026Official description of project workspaces, project knowledge, and project instructions.
U.S. Copyright OfficeCopyright Office Releases Part 2 of Artificial Intelligence ReportJanuary 29, 2025Official summary of the human-authorship conclusion for generative AI outputs.
Anthropic Privacy CenterIs my data used for model training? - ConsumersMarch 16, 2026Official consumer data-use conditions for model improvement, safety review, explicit opt-in, and incognito chats.
Federal Trade CommissionFTC Announces Crackdown on Deceptive AI Claims and SchemesSeptember 25, 2024FTC enforcement examples for deceptive AI claims, fake reviews, and unsupported performance promises.