Claude Code Misc Configurations

Collection of Claude Code Configurations

1. Meta-Prompter MCP And Slash commands

Meta-Prompter MCP

Location: .claude/mcp/meta-prompter/

A tiny meta‑prompt MCP server that grades prompts and returns JSON‑only analysis. Evaluates prompts across 8 dimensions (clarity, specificity, context, actionability, safety, testability, hallucination prevention, token efficiency) with weighted global scoring.

Key Features:

• Temperature-controlled evaluation (0 vs Claude Code's default 1)
• Machine-readable JSON output for agentic workflows
• Built-in result viewer with eval-viewer.html
• Support for multiple AI providers (Anthropic, OpenAI)

See meta-prompter README for detailed setup and usage instructions.

Commands

Location: .claude/commands/

Custom slash commands for Claude Code:

• /meta-prompter:eval - Quick prompt evaluation via Meta-Prompter MCP, returns JSON scores and analysis
• /meta-prompter:prep-run
  • Full workflow: evaluate → self-clarify or user-clarify based on chat context → execute with quality gates (global score ≥8)
• /debug-partner - AI-assisted debugging partner for systematic troubleshooting
  • Evidence-based investigation with code-centric traces
  • Progressive clarification and hypothesis testing
  • Includes specific file names, function names, and line numbers
  • Collaborative approach with verification at each step

2. Implement Plugin

Location: plugins/routine/

A Claude Code plugin that orchestrates JIRA ticket implementation end-to-end. Fetches ticket details, discovers domain knowledge, scans linked resources and designs, optimizes the prompt, then executes the task.

Skills: implement, jira-ticket-viewer, confluence-page-viewer, figma-reader, domain-discover, meta-prompter

See plugin README for setup and usage.

3. StatusLine

Location: .claude/statusline/

Context monitoring script that displays real-time usage:

• ctx_monitor.js
  • Shows context window usage percentage with color coding (green/yellow/red)
  • Tracks session ID and token consumption
  • Displays model name and usage statistics

Features:

• Real-time context window monitoring (0-200k tokens)
• Color-coded usage indicators
• Session tracking and identification
• Automatic synthetic message filtering