claude-code-misc

🥈Silver

A collection of Claude Code configurations for automating operations tasks. Includes a meta-prompt MCP server for grading and analyzing prompts, benefiting operations teams by streamlining workflows and improving prompt quality.

140Updated 2mo ago

Intermediate30min to implementautomation

Saves ~5 min per use

Quick InstallView Source

git clone https://github.com/delexw/claude-code-misc.git

Works with:

Claude

Overview

About This Skill

Claude Code Misc provides three integrated tools for operations automation: a Meta-Prompter MCP server that evaluates and grades prompts across clarity, specificity, context, actionability, safety, testability, hallucination prevention, and token efficiency; a debug-partner command for systematic, evidence-based troubleshooting with code-centric traces; and a StatusLine context monitor that displays real-time token usage and session tracking. The Meta-Prompter returns machine-readable JSON analysis with weighted global scoring, supporting multiple AI providers including Anthropic and OpenAI. This skill streamlines operations workflows by automating prompt quality assessment and providing collaborative debugging workflows with progressive hypothesis testing.

How to Use

Access the Meta-Prompter MCP server from .claude/mcp/meta-prompter/ with built-in HTML viewer for results. Use the /debug-partner command from .claude/commands/ for collaborative troubleshooting. Enable the StatusLine context monitor (ctx_monitor.js) from .claude/statusline/ to track real-time token usage.

Use Cases

Grade and optimize prompts before deployment using 8-dimension evaluation framework

Automate prompt quality analysis across teams with JSON-formatted results

Systematic debugging with evidence-based investigation and progressive clarification

Monitor context window usage in real-time with token consumption tracking

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/delexw/claude-code-misc

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Analyze the following prompt for [COMPANY] in the [INDUSTRY] sector and provide a detailed evaluation. The prompt is: '[PROMPT]'. Return your analysis in JSON format, including sections for clarity, specificity, potential biases, and suggested improvements.

Example Output

```json
{
  "analysis": {
    "clarity": "The prompt is generally clear but could benefit from more specific examples.",
    "specificity": "The prompt lacks specific details about the desired output format.",
    "potential_biases": "The prompt may unintentionally favor recent data over historical data.",
    "suggested_improvements": [
      "Include specific examples of desired outputs.",
      "Specify the time range for the data to be analyzed.",
      "Explicitly state the importance of both recent and historical data."
    ]
  },
  "score": 7.5
}
```