Documentation

Slop Sentry User Guide

Learn how to integrate the Slop Sentry into your workflow to ensure your AI-assisted writing sounds human, specific, and trustworthy.

Getting Started with Claude Desktop

The easiest way to use the Slop Sentry is via the Model Context Protocol (MCP) in Claude Desktop. This allows Claude to directly call our editorial tools while you chat.

  1. Get your API Key: Sign up for a plan to receive your private access token.
  2. Open Config: Open your Claude Desktop configuration file: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\\Claude\\claude_desktop_config.json (Windows).
  3. Add the Server: Paste the configuration snippet on the right into the mcpServers section.
  4. Restart Claude: Fully quit and restart Claude Desktop to enable the tools. 
{
  "mcpServers": {
    "slop-sentry": {
      "url": "https://mcp.slopsentry.ai",
      "headers": {
        "Authorization": "Bearer YOUR-API-KEY"
      }
    }
  }
}

Replace YOUR-API-KEY with your actual token.

Operating Modes

Two Ways to Protect Your Writing

The Slop Sentry offers two primary workflows depending on whether you want a quick fix or full manual control.

Recommended

Option A: The Writing Quality Sage

The writing_quality_sage tool is the fastest way to get polished results. It runs a 3-phase pipeline — Substance, Slop Sentry + Hokiness, AP Style — automatically rewriting your prose until every check reports CLEAN.

  • Automatic remediation across all four engines in one call
  • Iterative refinement with content hash deduplication
  • Cold-reader sub-agent evaluates substance for your target audience
  • Returns perfectly polished prose in seconds

Option B: Manual Pipeline

For maximum control, use individual tools step by step. This is ideal when you want to audit specific aspects of your writing independently.

  • 1. Detection: Use quality_check to run all four engines and get a unified violation report.
  • 2. Targeted Checks: Run substance_check, hokiness_check, or ap_style_check individually for focused audits.
  • 3. Re-Check: Run quality_check again with recheck: true to verify your manual edits.

Real-World Use

Conversational Sentry

You don't need to be a developer to use the Slop Sentry. Once the MCP server is connected, you can simply ask Claude to use the tools during your normal workflow.

General Proofing

"Run the writing_quality_sage on this draft. I want it to sound professional but human."

Claude will run the full 3-phase pipeline and return the clean, rewritten version.

Style-Specific

"Run the writing_quality_sage on my LinkedIn post with content_type: 'linkedin'."

Tailors the guard to enforce LinkedIn-specific rules like paragraph length limits and buzzword filters.

Custom Constraints

"Use the writing_quality_sage on this article, but set skip_ap_style and skip_hokiness."

Skip individual phases while keeping the core AI Tell guard and substance checks active.

Supported Content Types

Tailor the guard's behavior by specifying one of these types:

default article blog linkedin email proposal guide pptx image_text press_release

No MCP? No Problem.

Using SKILL.md as a System Prompt

If you are using an LLM that doesn't support MCP yet, or if you want to provide deeper context to your agent, you can use our SKILL.md file as a system prompt.

Simply copy the contents of the SKILL.md file (available in your dashboard) and paste it into your "System Instructions" or "Custom Instructions" field. This gives the AI the full logic of the Slop Sentry, even without the automated tools.

Developer Tools

The SLOPSENTRY CLI

For automated workflows, CI/CD pipelines, and local development, we provide a Python-based command-line interface.

Terminal — slopsentry
$ slopsentry check draft.md --type blog
❌ FAILED - 2 violations.
HYPOPHORA (1): Self-asked Q&A...
BUZZWORD (1): "delve"
$ slopsentry rewrite draft.md --provider remote
Job initiated. ID: 4f2a... Polling...
Content is now CLEAN.