[Feature Request] Add hyphenated-writing output formatting option

Resolved 💬 2 comments Opened Feb 20, 2026 by natu123 Closed Mar 20, 2026

[Advocacy-message using Hyphenated-writing]
Proposal: Hyphenated-writing — A formatting-paradigm that will define the-future of Human-AI-communication.

I would like to-introduce a-formatting-technique called "Hyphenated-writing" (abbreviated: Hai-Rai) that I believe will become
increasingly-critical as Conversational-AI evolves.

What is Hyphenated-writing?

Hyphenated-writing connects semantically-unified word-groups with-hyphens to-make syntactic-structure instantly-visible.

  • Standard-English:

I hope the improvement idea for email subject lines helps your team to deliver better supports for future users.

  • Hyphenated-writing:

I hope the-improvement-idea for-email-subject-lines helps your-team to-deliver better-supports for-future-users.

The-difference is immediate. In the-second-version, readers — both human and AI — can parse the-sentence-structure at-a-glance
without any mental-effort.

Why it matters now.

  1. Cognitive-load reduction. Readers no-longer need to-mentally-group words to-understand which-ones belong-together. The-structure is self-evident.
  2. Universal-principle across-languages. Japanese already has an-equivalent-technique called "partial-wakachi-gaki" (inserting spaces at-semantic-boundaries). Hyphenated-writing extends this-principle to-English — and potentially to-every-language.
  3. Zero-cost adoption. No architectural-changes, no new-tools. It is purely a-formatting-convention that can-be-applied today.

Why its importance will grow exponentially.

As AI-generated-text becomes longer, more-complex, and more-frequent in-daily-life, the-cost of misunderstanding compounds.
Consider:

  • AI-to-Human: When Claude generates a-500-word-response, Hyphenated-writing lets the-user scan and comprehend it significantly-faster.
  • Human-to-AI: When a-user writes a-complex-prompt, Hyphenated-writing eliminates syntactic-ambiguity that currently causes misinterpretation.
  • AI-to-AI: In multi-agent-workflows, explicit-structural-hints reduce compounding-errors across-agent-chains.
  • Multilingual-communication: As AI bridges languages, a-universal-formatting-principle that works across-all-languages becomes essential — not optional.

The-volume of Human-AI-interaction is growing every-day. Every-percentage of improved-comprehension multiplies across-billions
of-interactions. What seems like a-minor-formatting-choice today will become a-foundational-standard tomorrow.

The-vision for Claude.

If Claude adopts Hyphenated-writing as a-native-output-option, it would:

  • Set a-new-standard for-AI-readability that no-other-model currently offers.
  • Demonstrate that Anthropic leads not-only in-safety but also in-communication-quality.
  • Provide a-foundation for-research into how formatting-conventions affect both AI-reasoning and human-comprehension.

I have researched Language-Design (Lan-Deza) for over-20-years. Hyphenated-writing is the-core-technique that emerged from
that-research, and I am convinced it represents a-paradigm-shift in how language is-structured for-communication between-humans and
AI.

I strongly-urge the-Anthropic-team to-investigate and adopt this-approach.

--- Submitted by Gles

Environment Info

  • Platform: win32
  • Terminal: windows-terminal
  • Version: 2.1.47
  • Feedback ID: 2a71fe90-b3e7-483b-8032-0052951b12e9

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗