berelevant-ai/slopless

What it solves

It eliminates "AI-slop" prose—the hollow, repetitive, and predictable writing patterns typically generated by LLMs—to ensure AI-generated text reads as human and authentic as possible.

How it works

Slopless is a deterministic linter that flags specific "LLM tells" such as hollow framing, fake contrasts, hedging, and vacuous closers. Unlike many AI tools, it does not use a model or API keys; it uses a set of rules to identify problematic rhetoric and structure. It can be used as a standalone zero-config CLI or as a preset for the textlint ecosystem. It also provides "skills" that can be installed for AI agents (like Claude Code or Codex), allowing them to run the linter, identify errors, and iterate on their own writing until the text passes.

Who it’s for

It is designed for developers and writers using AI writing agents who want to automate the process of refining AI-generated content to remove robotic patterns.

Highlights

• Deterministic Linting: Uses a rule-based approach rather than another LLM to flag prose issues.
• Agent Integration: Includes specific installation commands to give writing agents (Claude, Codex) the ability to self-correct their output.
• Flexible Deployment: Works as a standalone CLI or as a textlint preset.
• Zero-Config: The CLI requires no separate installation of textlint or configuration files to start flagging slop.