en-scorelisted

# English Text Quality Score Score the text in $ARGUMENTS. With no arguments, score the most recent English text in the conversation. If $ARGUMENTS names a file or a glob, read those files. ## Where the corpus lives Find a directory named `references` whose parent directory is named `en-text`. It sits next to this skill, typically at `../en-text/references/`. Read `scoring.md` first; it governs everything below. Read `slop.md` for the density table, `method.md` before you trust a count, and `clarity.md` and `usage.md` as needed to identify what you are counting. If the corpus is missing, say so and stop. A number produced from memory is not an en-text score. ## Procedure 1. **Count.** Words of running prose, excluding code blocks, tables, front matter, and quoted material. State the count. 2. **Identify the genre and weights.** Default weights are in `scoring.md`. If the genre shifts them, say which weights you used and why. 3. **Score each dimension** against its anchor table. Count instances before assigning a number; do not score by impression. 4. **Combine** with the weights and report to one decimal. Show the arithmetic to yourself before printing: the total must follow from the sub-scores. 5. **Name the three findings that cost the most points**, each with its rule ID. ## Output ``` Reviewed as: engineering blog post, US spelling, default weights. 840 words. Score: 6.1 / 10 Sentence clarity 5.5 ██████░░░░ Flow and cohesion 7.0 ██