arielperez82

reader-clarity

Evaluate and improve editorial content readability using Flesch-Kincaid

arielperez82 0 Updated 3mo ago

Resources

1
GitHub

Install

npx skillscat add arielperez82/agents-and-skills/reader-clarity

Install via the SkillsCat registry.

SKILL.md

Reader Clarity

Evaluate and improve editorial content readability for general audiences.

Overview

Newsletter readers are scanning, not studying. They read on phones during commutes, in email clients between meetings, and in bed before sleep. Content must be immediately clear — no re-reading required, no jargon lookups, no "wait, what does that refer to?" moments. This skill provides a systematic framework for assessing and improving readability.

Core Value: Every reader should understand every sentence on first read, regardless of their domain expertise.

Core Capabilities

  • Readability Scoring — Flesch-Kincaid grade level targeting 8-10 for general newsletter audiences
  • Context Budget — Limit assumed reader knowledge per story (max N context credits per article)
  • Jargon Detection — Domain-specific jargon checklist with plain-language alternatives
  • Rewriting Patterns — Systematic transforms: jargon-to-plain, complex-to-simple, passive-to-active
  • Structural Clarity — Buried lede detection, unclear antecedents, transition gaps

Quick Start

  1. Read the article to review
  2. Estimate Flesch-Kincaid grade level (target: 8-10)
  3. Count assumed-context credits (target: max 5 per story)
  4. Scan against jargon checklist at references/jargon-checklist.md
  5. Apply rewriting patterns to flagged passages

Key Workflows

1. Full Readability Assessment

  1. Read the article — Note passages that require re-reading
  2. Estimate readability score — Flesch-Kincaid grade level:
    • Grade 8-10 = target for general newsletters
    • Grade 6-8 = very accessible (good for broad audiences)
    • Grade 10-12 = too complex for scanning (acceptable for niche/expert audiences)
    • Grade 12+ = academic level (not suitable for newsletters)
  3. Count context credits — Each of these costs 1 credit:
    • Unexplained acronym or abbreviation
    • Reference to a prior event without brief context
    • Industry-specific term used without definition
    • Insider reference (assumes reader follows the topic closely)
    • Implicit causal chain (assumes reader connects A→B→C)
  4. Scan for jargon — Check against domain-specific jargon checklist
  5. Check sentence structure:
    • Passive voice constructions
    • Sentences over 25 words
    • Unclear antecedents ("it," "this," "they" without clear referent)
    • Buried lede (key information not in first 2 sentences)
    • Missing transitions between paragraphs
  6. Produce readability report with scores, flagged passages, and rewrites

2. Quick Clarity Pass

  1. Focus on the 3 highest-impact checks only:
    • Jargon (scan against checklist)
    • Context credits (count assumed knowledge)
    • Buried lede (is the key point in the first 2 sentences?)
  2. Flag and provide rewrites for any issues found

Rewriting Patterns

Jargon to Plain

Jargon Plain
"monetize" "make money from"
"leverage" "use"
"synergy" "combined benefit" or [just describe the specific benefit]
"pivot" "change direction" or "shift focus"
[See full list in jargon-checklist.md]

Complex to Simple

Complex Simple
"notwithstanding the aforementioned considerations" "despite this"
"it is important to note that" [delete — just state the fact]
"in the event that" "if"
"at this point in time" "now"
"due to the fact that" "because"

Passive to Active

Passive Active
"The decision was made by the Fed" "The Fed decided"
"It was reported that earnings fell" "The company reported falling earnings"
"The policy is expected to be implemented" "[Actor] plans to implement the policy"

Best Practices

  • Fix jargon and context budget issues first — these are the biggest barriers to comprehension
  • Readability scores are guides, not rules — a score of 11 with zero jargon may be fine
  • When simplifying, never lose precision — "The Fed raised rates by 0.25%" is better than "The Fed raised rates a little"
  • Context budget resets per story in a multi-story newsletter — each story must stand alone
  • The first sentence of each story should be understandable by anyone

Reference Guides

  • jargon-checklist.md — Domain-specific jargon organized by field with plain-language alternatives

Integration

Core skill for reader-clarity-reviewer agent. Also useful for editorial-writer during drafting (awareness mode).