Use a source hierarchy
For Claude-specific facts, primary sources are Anthropic and Claude properties: platform docs, help-center articles, privacy pages, release notes, and terms. For legal and regulatory claims, primary sources are government publications and official notices.
Personal workflow posts can be useful inspiration, but they should not be the basis for current claims about availability, privacy, copyright, or policy. Those claims change and need owners.
This site cites primary sources wherever it makes factual claims about Claude features, data handling, copyright registration, or deceptive AI claims.
Claude product and prompting sources
Use Anthropic platform docs for prompt engineering and model-platform concepts. Use Claude Help Center articles for consumer and app-surface features such as projects, artifacts, research, skills, and Microsoft Office add-ins.
When a help-center article has a publish or update date, cite it. When it does not, include your access date in your own notes.
Model, pricing, and provenance sources
Model facts decay faster than anything else a writing guide asserts, so they get their own tier. For which model to use, cite the choosing-a-model page, which carries the selection matrix and the statement that most workloads start with Claude Opus 5. For model IDs, context windows, output limits, and knowledge cutoffs, cite the models overview and comparison table. For rates, cite the pricing page on the day you publish. Launch announcements for Claude Opus 5 and Claude Fable 5.1 are the right citations for release dates and positioning.
For provenance questions, the primary source is Anthropic’s watermarking explainer, which covers the method, its limits, the detection API, and C2PA content credentials on generated image files. Do not cite third-party AI-detection vendors for claims about Claude watermarking; they use a different technique and do not hold Anthropic’s key.
For export surfaces, cite the Help Center article on creating and editing files for .docx, .pptx, .xlsx, and PDF output, and the Claude Design announcement for slides, one-pagers, and deck export. For memory behavior across chat and Claude Cowork, cite the Claude blog post on memory.
Privacy and retention sources
Use the Anthropic Privacy Center for data-use and retention claims. Separate consumer products from commercial products because their defaults and controls differ.
For sensitive writing, cite the exact policy page that applies to the product surface and plan being used.
Copyright and marketing-law sources
Use the U.S. Copyright Office AI hub, its Part 2 copyrightability summary, and the Federal Register registration guidance for AI-assisted authorship. Use FTC releases and business guidance for deceptive AI claims and fake review risks.
This site does not provide legal advice. It provides source-grounded workflow guidance so writers know which questions to raise before publication.
FAQ
Does Claude Writes cite blogs?
Not for current feature, privacy, legal, or policy claims. Blogs may inspire workflow questions, but primary sources carry factual claims.
Why are source dates included?
Claude features and policies change. Dates help readers decide whether a claim needs refreshing.
Is this site affiliated with Anthropic?
No. Claude Writes is an independent educational publication and is not affiliated with Anthropic.
Primary sources
Official overview of Claude as a language, reasoning, analysis, and coding platform.
Anthropic DocsPrompting best practicesReviewed 2026-09-19Official guidance on clarity, context, examples, XML tags, roles, and long-context prompt structure.
Anthropic DocsChoosing a modelReviewed 2026-09-19Official model-selection matrix and the guidance that most workloads start with Claude Opus 5.
Anthropic DocsModels overview and comparison tableReviewed 2026-09-19Official model IDs, context windows, output limits, and knowledge cutoffs for the current lineup.
Anthropic DocsPricingReviewed 2026-09-19Official per-token pricing for the current lineup, including the Claude Sonnet 5 rate.
AnthropicClaude Opus 5July 24, 2026Official Opus 5 launch announcement, including its role as the default model on Claude Max.
AnthropicClaude Fable 5.1 and Claude Mythos 5.1September 1, 2026Official Fable 5.1 launch announcement for demanding reasoning and long-horizon agentic work.
Claude Help CenterWhat are projects?March 16, 2026Official description of project workspaces, project knowledge, and project instructions.
Claude Help CenterUnderstanding Claude's personalization featuresMay 28, 2026Official explanation of account instructions, project instructions, and styles.
Claude BlogClaude's memory works everywhere, and you decide what's in itSeptember 2026Official announcement that chat memory and Cowork memory are the same, with topic-level review, editing, deletion, and sensitive topics off by default.
Claude Help CenterWhat are skills?June 1, 2026Official definition of skills as folders of instructions, scripts, and resources for repeatable tasks.
Claude Help CenterWhat are artifacts and how do I use them?Updated 2026Official description of artifacts as a separate surface for substantial standalone content.
Claude Help CenterUse research on ClaudeJune 2026Official note that Claude Research works with web search and returns checkable citations.
Claude Help CenterCreate and edit files with ClaudeAugust 6, 2026Official documentation for creating .docx, .pptx, .xlsx, and PDF files, with the note that outputs may need review and refinement.
AnthropicIntroducing Claude Design by Anthropic LabsApril 17, 2026Official description of Claude Design for slides, one-pagers, and decks, including export to PDF, PPTX, Canva, or HTML.
AnthropicHow Claude's text watermark worksAugust 14, 2026 (updated September 1, 2026)Official explanation of Claude text watermarking: SynthID-Text method, EU AI Act compliance, no identifying information, and what a watermark does and does not prove.
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.
Anthropic Privacy CenterIs my data used for model training? - Commercial customersMarch 16, 2026Official commercial-product default that inputs and outputs are not used for training unless feedback or permission is given.
U.S. Copyright OfficeCopyright and Artificial IntelligenceReviewed 2026-09-19Official hub for the Copyright Office AI reports and public-comment process.
U.S. Copyright OfficeCopyright Office Releases Part 2 of Artificial Intelligence ReportJanuary 29, 2025Official summary of the human-authorship conclusion for generative AI outputs.
Federal RegisterCopyright Registration Guidance: Works Containing Material Generated by AIMarch 16, 2023Official registration guidance requiring disclosure of AI-generated material and identification of human-authored contributions.
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.