prose-polish_skill

This skill analyzes and elevates prose quality across craft, coherence, authority, purpose, and voice to produce clear, credible, and engaging writing.
  • Python

20

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill leegonzales/aiskills --skill prose-polish

  • CHANGELOG.md489 B
  • README.md1.3 KB
  • SKILL.md14.3 KB

Overview

This skill evaluates and elevates writing effectiveness across craft, coherence, authority, purpose, and voice with genre-calibrated thresholds. It focuses on making writing genuinely better—clearer logic, earned expertise, purposeful stakes, and a distinctive voice—rather than simply masking AI origin. Use it to diagnose drafts, guide rewrites, or teach practical writing principles.

How this skill works

The skill first detects genre and applies genre-specific thresholds, then runs a six-dimension analysis (Craft, Coherence, Authority, Purpose, Voice, Effectiveness). For remediation it performs a two-phase process: Phase 1 fixes structure and authority signals, Phase 2 refines rhythm, commitment, and voice. For generation it builds genre-calibrated constraints, generates text, and self-verifies against the six dimensions.

When to use it

  • You have a draft and need a diagnostic quality profile
  • You want a phased rewrite that fixes logic before style
  • Generate new content with guaranteed genre-aware quality checks
  • Teach or practice writing using targeted dimension exercises
  • Tune hedging and authority signals for credibility

Best practices

  • Detect genre before scoring; thresholds change by genre
  • Run Phase 1 (coherence + authority) before any stylistic edits
  • Classify hedges—preserve epistemic honesty, remove cowardly hedges
  • Use conservative/moderate/aggressive depth controls to protect voice
  • Prefer radical simplicity when a detail doesn’t advance the argument

Example use cases

  • Analyze a product announcement to ensure clear stakes and ownership
  • Elevate an academic draft: improve contribution statement and syntheses
  • Rewrite marketing copy to keep personality but add demonstrable authority
  • Generate technical docs that explain why decisions were made, not just how
  • Teach students to spot decorative specificity and repair logical gaps

FAQ

It measures signaling—style, structure, and markers of authority—but does not fact-check claims. Scores reflect whether expertise is demonstrated, not whether facts are true.

How aggressive should remediation be?

Default is Moderate (both phases, light Phase 2). Use Conservative to preserve voice and fix logic only, or Aggressive for a full rewrite.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational