AI Readability vs Readability

AI Readability is a web checker for English text aimed at writers, editors, and content teams. It measures how readily people can read text and how easily AI systems can parse and reuse it. The checker returns a score from 0 to 100 covering Sentence Clarity, Structural Organization, Entity Consistency, and Answer Readiness. It also offers a short illustrative rewrite, rather than a complete edit. Scoring runs entirely in the browser, so pasted text is not sent to a server during the check. The site says the tool uses transparent heuristics without a model call, server, or crawl, and requires no signup. English is the only supported language. Selecting “Continue in Textorum” sends the text to Textorum through the URL. The maker describes the score as a proprietary diagnostic framework, not a validated scientific instrument, and says no readability tool can promise citations or rankings. The free plan costs nothing.

Readability analyzes Markdown files and reports measures of reading complexity. It provides grade level, reading ease and estimated reading time calculated at 200 words per minute. The listed grade-level formulas are Flesch-Kincaid, Gunning Fog, Coleman-Liau, SMOG and ARI. It also measures word and sentence metrics and can detect or require MkDocs-style admonitions. Results can be produced as a table, Markdown, JSON, summary or report. Configurable thresholds can fail CI checks when documentation exceeds them, and a.readability.yml file supports path-specific overrides. The project documents a Go command-line tool, Docker, a GitHub Action and pre-commit hooks, alongside CI/CD examples for GitLab CI, CircleCI and Jenkins. Pre-built binaries are listed for Linux and macOS in amd64 and arm64, and Windows in amd64. The project identifies its license as MIT and lists no paid plans. Its documentation cautions that scores measure word and sentence complexity, not accuracy, completeness, context or tone.