deslop

Detect AI-generated slop in comments and docstrings with suggested rewrites.

Updated May 2, 2026

Search Tags:#code review #grep #repo hygiene #comment analysis #documentation integrity #docstring cleanup #AI-generated text

Ready to run

npx skills add https://github.com/ajbarea/techne --skill deslop-ajbarea

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deslop
Source: https://github.com/ajbarea/techne/tree/main/plugins/techne/skills/deslop
Command: npx skills add https://github.com/ajbarea/techne --skill deslop-ajbarea

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes low-value, AI-generated or rot-prone commentary from comments and docstrings so your codebase stays readable and trustworthy over time.

Core Features & Use Cases

Slop detection across comments/docstrings: Flags temporal markers, self-referential AI framing, narrative restatements, marketing padding, hedging, and unmeasured performance claims.
Grounded handling for claim-bearing docstrings: Routes candidates that require implementation verification to the sibling /techne:reslop for code-grounded rewrites.
Safe, scoped auditing: Scans the whole repo with skip-path conventions or narrows to a specific path/pattern while protecting reviewed sections and critical headers/directives.

Quick Start

Run deslop on the current repository to generate a grouped list of comment/docstring findings with suggested deletions or trims.

Frequently Asked Questions about deslop

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

FAQPage Schema

How do I remove AI-generated comments and docstrings from my repository?▼

To remove AI-generated comments and docstrings, run a whole-repo or path-scoped audit that flags temporal markers, hedging, and narrative padding, then applies tightened rewrites or deletions to improve repo hygiene.

What is AI slop in code comments and docstrings?▼

AI slop in code comments refers to low-value, rot-prone text like self-referential AI framing, marketing padding, unmeasured performance claims, and narrative restatements that reduce codebase readability over time.

How do I clean up docstrings that contain unverified performance claims?▼

To clean up docstrings with unverified performance claims, route the flagged candidates to code-grounded handling for verification and rewriting, ensuring the documentation accurately reflects the implementation.

Can I audit specific paths in my repo for documentation integrity without scanning everything?▼

Yes, you can narrow your documentation integrity audit to a specific path or pattern using skip-path conventions, which protects reviewed sections and critical headers while scanning targeted source files.

When should I clean up AI-generated text during the code review process?▼

You should clean up AI-generated text during PR review, after merges, or when incorporating code from other assistants to prevent low-value commentary from accumulating and degrading repository hygiene.