deslop

Rewrite AI-sounding text into grounded prose with diagnostic output.

2|Updated May 21, 2026
One-click install
npx skills add https://github.com/howells/skills --skill deslop-howells
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deslop
Source: https://github.com/howells/skills/tree/main/deslop
Command: npx skills add https://github.com/howells/skills --skill deslop-howells

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you spot and remove common AI-writing tells so your text sounds more specific, human, and credible instead of generic or detector-triggering.

Core Features & Use Cases

  • Diagnose AI-writing tells: Detect high-signal clusters like inflated significance, vague attribution, fake contrasts, assistant leakage, and citation/markup artifacts.
  • Rewrite with editorial guardrails: Preserve meaning while replacing synthetic cadence with direct, verifiable statements and native markup.
  • Handle multiple input types: Review inline text, local files, or fetched URLs (when browsing is available), and limit diagnosis length when the user only requests a rewrite.

Quick Start

Use deslop to audit this draft for AI-writing tells and rewrite it into a more grounded voice.

Frequently Asked Questions about deslop

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I rewrite AI-sounding prose into more grounded copy?

To rewrite AI-sounding prose, you need to detect high-signal clusters like inflated significance and vague attribution, then replace synthetic cadence with direct, verifiable statements while preserving the original meaning and markup.

What are common AI-writing tells I should look for in a draft review?

Common AI-writing tells include inflated significance, vague attribution, fake contrasts, assistant leakage, and citation or markup artifacts that make prose sound generic and trigger AI detectors.

Can I clean up AI-writing artifacts in local files and fetched URLs?

Yes, you can clean up AI-writing artifacts across inline text, local files, and fetched URLs when browsing is available, applying taxonomy-based diagnostics to correct citation and markup issues.

How do I fix citation and markup artifacts left by AI text generation?

Fix citation and markup artifacts by detecting them as a specific AI-writing cluster, then rewriting the affected text into native markup while preserving the host document's structure and meaning.

Does rewriting AI-generated text preserve the original meaning and host markup?

Yes, the rewriting process uses editorial guardrails to preserve the original meaning and host markup, replacing only the synthetic cadence with specific, human-sounding, and verifiable statements.

What is the output format for an AI prose review?

An AI prose review outputs a structured format containing Observed tells, Why they matter, the Rewrite, and Residual risk, ensuring transparent diagnostics and grounded copy correction.