create-meta-prompts

Create prompts for Claude-to-Claude multi-stage workflows with XML metadata.

Updated Apr 30, 2026
One-click install
npx skills add https://github.com/emptist/psypi --skill create-meta-prompts-emptist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-meta-prompts
Source: https://github.com/emptist/psypi/tree/main/ppi_skills/create-meta-prompts
Command: npx skills add https://github.com/emptist/psypi --skill create-meta-prompts-emptist

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill addresses the challenge of manually crafting prompts that produce structured outputs for subsequent prompts, facilitating efficient multi-stage workflows in Claude-to-Claude communication.

Core Features & Use Cases

  • Multi-Stage Workflow Support: Handles tasks like research, planning, and implementation with structured XML metadata.
  • Automatic Dependency Detection: Scans for existing research/plan files and suggests relevant references.
  • Clear Provenance and Dependency: Each prompt is in its own folder with clear output artifacts and automatic dependency detection.
  • Use Case: For instance, if you're working on a new application, this skill can assist in creating a structured workflow from initial research to final implementation.

Quick Start

To create a meta-prompt for a new research task, use: /create-meta-prompt research authentication options for the app

Frequently Asked Questions about create-meta-prompts

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

FAQPage Schema
How do I create structured multi-stage prompts for Claude-to-Claude workflows?

To create structured multi-stage prompts for Claude-to-Claude workflows, you generate prompts that handle research, planning, and implementation stages using XML metadata. This ensures outputs are efficiently parsed by subsequent prompts in the pipeline.

What is the best way to automate prompt generation for research and implementation stages?

Automating prompt generation for research and implementation stages is best handled by creating meta-prompts that output structured XML metadata. This allows each stage to pass formatted data directly to the next Claude instance automatically.

How does automatic dependency detection work when generating workflow prompts?

Automatic dependency detection works by scanning your environment for existing research or plan files. It then suggests relevant references to include in the newly generated prompt, ensuring clear provenance and output artifacts for your workflow.

Do I need XML metadata to parse structured outputs from Claude?

Yes, you need XML metadata to parse structured outputs from Claude effectively. The skill requires XML metadata to structure the outputs of research, planning, and implementation prompts so subsequent stages can parse them without manual intervention.

Can I use this skill to build a workflow from initial research to final implementation?

Yes, you can use this skill to build a workflow from initial research to final implementation. It creates structured prompts for each phase, placing them in distinct folders with clear output artifacts to support full application development pipelines.