epic-meta-prompting

Generate research and planning prompts for epic workflows with isolated sub-agents.

91|10|Updated Oct 4, 2025
One-click install
npx skills add https://github.com/marcusgoll/Spec-Flow --skill epic-meta-prompting
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: epic-meta-prompting
Source: https://github.com/marcusgoll/Spec-Flow/tree/main/.claude/skills/epic-meta-prompting
Command: npx skills add https://github.com/marcusgoll/Spec-Flow --skill epic-meta-prompting

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Generates research and planning prompts for epic workflows via isolated sub-agents to avoid context contamination.

Core Features & Use Cases

  • /create-prompt for research
  • /run-prompt to execute and store outputs
  • References to research and planning artifacts
  • Copy-ready handoff prompts for next phase

Quick Start

Create and run separate prompts for research and planning to keep context clean.

Frequently Asked Questions about epic-meta-prompting

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

FAQPage Schema
How do I use meta-prompting to keep research and planning separate in epic workflows?

Meta-prompting isolates sub-agents to avoid context contamination during epic research and planning phases. Create separate prompts for each phase using /create-prompt, execute them via /run-prompt, and store outputs with XML metadata in .prompts directories to maintain clean context boundaries.

What's the best way to generate research prompts for epic projects?

Use /create-prompt to generate research prompts that detect purpose and solicit contextual questions. Reference project documentation as needed, then execute prompts with /run-prompt to auto-store findings and artifacts, producing research.md with structured outputs.

Can I automate prompt execution and artifact export during epic planning?

Yes. Meta-prompting automates prompt execution via /run-prompt and exports findings with XML metadata extraction to .prompts directories. This supports the full epic lifecycle from research through planning with copy-ready handoff prompts for downstream phases.

How do sub-agents prevent context contamination in prompt workflows?

Isolated sub-agents compartmentalize each phase of epic workflows, ensuring research findings don't bleed into planning decisions. This separation maintains reasoning clarity and supports metadata extraction for commits and phase transitions.

What outputs does meta-prompting produce for epic handoffs?

Meta-prompting generates research.md and plan.md artifacts with XML metadata, stored in .prompts directories. These outputs include copy-ready handoff prompts structured for the next phase and support commit-ready metadata extraction.