Skip to main content
skill CX Codex

tighten-prose — Claude Code skill: editing prose for AI tells

tighten-prose strips the habits readers spot as machine-written — word clusters, inflated importance, phantom citations — and rewrites the flagged passages while keeping the author's voice.

MIT ★ 114 updated 2026-08-11

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add nicknisi/claude-plugins
/plugin install content@nicknisi
Open on GitHub

What tighten-prose does

Point this skill at finished prose and it removes the habits that read as machine-written: clusters of overused words, importance inflated without new facts, participial clauses tacked on for depth that is not there, attributions to experts nobody named, connectors from a school essay. It judges density. One stray “crucial” passes; three of them across two paragraphs gets the section rewritten.

The pass opens the bundled catalog in references/ai-tells.md, then fetches the current “Words to watch” list from Wikipedia over the API, since the tells drift as models change. If that request fails, the static catalog carries the work. Findings come back as clusters rather than a tally of every match, and the flagged passages get rewritten with meaning and voice intact.

Structure is where it goes blind. A draft that ends every section on an aphorism, or leans on the same antithesis over and over, comes through clean. Run it last, and weight it least among your quality gates.

What it helps with

  • strip the tells that make a draft read as generated
  • scan an article for AI vocabulary clusters before publishing
  • rewrite a paragraph of inflated importance down to its facts
  • clean phantom citations out of documentation
  • add a final editing pass on top of another skill's draft

Example prompts for the agent

  • Tighten this up — cut whatever makes it read like AI.
  • Check the draft for AI tells and rewrite the paragraphs you flag.
  • Run a writing pass over docs/guide.md and keep my voice.

How to install tighten-prose

1Plugin marketplace

/plugin marketplace add nicknisi/claude-plugins
/plugin install content@nicknisi

This installs the whole content plugin with all its skills

2Skills CLI (Vercel)

npx skills add nicknisi/claude-plugins --skill tighten-prose -a claude-code

3Manually: put the files in place

~/.claude/skills/tighten-prose/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/tighten-prose/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about tighten-prose

How do I install tighten-prose in Claude Code?

Add the marketplace with /plugin marketplace add nicknisi/claude-plugins, then install the plugin: /plugin install content@nicknisi. The skill ships inside the content plugin.

What does tighten-prose do?

It scans prose for the pattern categories that show up disproportionately in AI output, reports where the density runs above noise, and rewrites the marked sections without changing their meaning.

What can't it catch?

Structure. Vocabulary and sentence-level habits are visible to it, while a draft that closes every section the same way slips past.

Source text

SKILL.md · MIT
---
name: tighten-prose
description: This skill should be used when the user asks to "tighten this up", "tighten the prose", "clean up the writing", "check for AI tells", "make this sound more human", "de-slop this", "run a writing pass", or when completing a writing task that should not read like AI output. Scans prose for statistically overrepresented AI writing patterns and rewrites flagged sections.
---

# Tighten

A final editing pass that catches and removes AI-detectable writing patterns. Not a detector — an editor. The output reads like a human wrote it because, after this pass, a human shaped every sentence.

## What this pass cannot see

**This skill scans vocabulary and sentence-level habits. It is blind to structure,
and structure is where modern AI prose gives itself away.**

A ghost-written blog post once passed this gate with zero findings — no AI
vocabulary, em-dash rate *below* the author's own baseline — while ending eight of
eight sections on an aphorism, using six "Not X. That's Y." antitheses, and
asserting nine invented claims about the author's inner life. Readers called it
slop immediately. The gate had certified its own blind spot.

If you are tightening a draft written *for* someone else in their voice, run a
corpus-calibrated structural check first — see
`plugins/content/skills/blog-post-writer/scripts/slopcheck.py`. Two rules that pass
list can't encode:

- **Slop is a rate, not a presence.** Real writers use every device on the list
  below. The tell is one firing on every section instead of once a post. Thresholds
  have to be measured against the author's real corpus.
- **Check both directions.** A metric sitting far *below* an author's baseline means
  someone wrote to the metric. That's as suspicious as excess.

Run this skill last and weight it least.

## When to Run

- After completing any prose-heavy output (blog posts, documentation, emails, talks)
- When another skill requests a tighten pass as a quality gate
- When the user asks to clean up or improve writing
- Before shipping anything where "sounds like AI" would undermine credibility

## Process

### 1. Load the Tells Reference

Read `references/ai-tells.md` for the full catalog of patterns, organized by category with examples and fix strategies.

### 2. Fetch Live Patterns

AI writing tells evolve as models change. Fetch the current vocabulary list from Wikipedia:

```
https://en.wikipedia.org/w/api.php?action=parse&page=Wikipedia:Signs_of_AI_writing&prop=wikitext&format=json
```

Extract the "Words to watch" lists and AI vocabulary entries. Newer models drop old tells and develop new ones — the live source catches drift the static reference cannot.

If the fetch fails, the static reference is sufficient. Proceed without it.

### 3. Scan for Pattern Density

Read the full text and scan across all categories:

| Category                 | Signal                                                      |
| ------------------------ | ----------------------------------------------------------- |
| Vocabulary clusters      | 3+ AI-overused words in a section                           |
| Inflated significance    | Phrases that inflate importance without adding information  |
| Superficial -ing clauses | Participial phrases tacked on as fake depth                 |
| Promotional language     | Travel-brochure / sales-deck vocabulary                     |
| Vague authority          | Phantom expert attributions ("Industry reports suggest...") |
| Formulaic transitions    | Five-paragraph-essay connectors (moreover, furthermore)     |
| Negative parallelism     | Overuse of "not X, it's Y" construction                     |
| Rule of three            | Every list having exactly three items                       |
| Copula avoidance         | "Serves as" instead of "is"                                 |
| Elegant variation        | Thesaurus cycling to avoid repeating a concrete noun        |
| Em dash density          | Em dashes at 2-3x human rate                                |
| Formatting tells         | Mechanical boldface, title case headings, emoji decoration  |

**The signal is density, not individual hits.** One "crucial" is fine. A cluster of three AI vocabulary words in two paragraphs is a rewrite.

### 4. Report Findings

Present a concise summary:

- Which categories showed pattern density above noise level
- Specific examples quoted from the text
- Severity: how many sections need work vs. isolated hits

Do not list every individual word match. Report clusters and structural patterns.

### 5. Rewrite Flagged Sections

For each flagged section, rewrite to eliminate tells while preserving meaning.

**Fix strategies:**

- **Inflated significance** — Say what actually happened. Specific facts beat vague importance.
- **Superficial -ing** — If the clause adds no new information, cut it entirely.
- **Copula avoidance** — If a fancy verb replaces "is" without reason, use "is."
- **Formulaic transitions** — Use casual transitions or drop them. "Here's the thing:" beats "Moreover."
- **Elegant variation** — Repeat the concrete noun. "Vim" twice is better than "vim... the editor... the tool."
- **Rule of three** — Vary list lengths. Two items is fine. Four is fine. One is fine.
- **Vague authority** — Name the person or drop the attribution.
- **Promotional language** — Replace with specifics. What exactly makes it good?
- **Em dashes** — Keep the ones that earn their place. Replace the rest with commas or periods.

**Constraints:**

- Preserve the original meaning and information content
- Maintain the existing voice and tone — do not flatten personality
- Keep legitimate rhetorical devices. A single "not X, it's Y" used for emphasis is a deliberate choice, not a tell. The tell is when every other paragraph uses it.
- When a pattern appears intentional and isolated, leave it

### 6. Present the Tightened Version

If only a few sections changed, present before/after pairs for the changed passages. If the whole piece needed work, present the full rewrite.

Do not explain every change. The writing should speak for itself. Mention only changes where the reasoning is non-obvious.

## Composability

Designed to work standalone or as a step in other writing workflows.

**Standalone:** Point at text and say "tighten this up."

**Composed:** Another skill (like blog-post-writer) calls for a tighten pass as its final quality gate. When composed, respect any voice calibration the parent skill has already established. Tighten removes autopilot patterns, not personality.

## Bundled Resources

- **`references/ai-tells.md`** — Full catalog of AI writing tells with examples, fix strategies, and density guidance. Adapted from Wikipedia's "Signs of AI writing."

© Nick Nisi · MIT · nicknisi/claude-plugins

Similar skills

All skills for Writing and docs →

From the same set18

The whole set →
thermo-nuclear-code-quality-review thermo-nuclear-code-quality-review conducts a ruthless maintainability audit on feature branches. Flags tangled logic, oversized files, and opportunities to cut structural bloat. CX Codex ★ 114repo stars explain-diff The explain-diff skill turns a diff, PR, branch, or commit into an HTML guide with system context, diagrams, a code walkthrough, and a quiz. Use it to understand a change or teach it to someone else. CX Codex +1 ★ 114repo stars code-simplifier code-simplifier refactors complex functions, cuts redundancy, and removes AI slop without touching public APIs or changing runtime behavior. CX Codex ★ 114repo stars git-committer The git-committer subagent reviews repository changes, runs available checks, and creates commits with semantic messages. Use it to hand off the commit workflow to Claude Code. CX Codex ★ 114repo stars blog-post-writer blog-post-writer converts raw notes, git commits, and logs into a fact-checked, flat blog post draft with explicit author question markers. CX Codex +1 ★ 114repo stars conference-talk-builder conference-talk-builder turns rough notes, transcripts, or a topic into a structured tech talk script with a slide-by-slide plan. Use it to shape ideas into a narrative. CX Codex +1 ★ 114repo stars youtube-notes youtube-notes is a Claude Code skill that turns YouTube captions into chapter-based notes with clickable timestamps. It activates on a YouTube link or a request to analyze a video. CX Codex +1 ★ 114repo stars security-auditor This subagent checks a codebase for critical security flaws through trust-boundary and invariant analysis. Use it for an explicit security audit or vulnerability assessment. CX Codex ★ 114repo stars

Project sponsors

Companies that keep this analytics open