create-meta-prompts
Generate meta-prompts for Claude-to-Claude pipelines and multi-stage workflows. Use when creating optimized prompts for complex workflows, delegating to subagents, or managing multi-stage execution. Includes XML output structuring, metadata injection, and chain provenance. Not for simple single-turn prompts, manual messaging, or non-Claude integrations. Keywords: prompt chain, delegation, multi-stage, workflow orchestration, Claude-to-Claude.
npx skills add majiayu000/claude-skill-registry --skill create-meta-prompts-git-fg-meta-plugin-manager --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Generates meta-prompts intended to drive Claude-to-Claude pipelines and multi-stage workflows. It creates a prompt folder in a structured workspace, produces an executive SUMMARY.md, and maintains chain provenance for multi-step execution. It supports generating prompts in a chained, dependency-aware fashion and provides templates and patterns for workflow orchestration, including workflow phases, purpose inference, and SUMMARY.md structure.
How it works
- Objective: Generate meta-prompts for Claude-to-Claude pipelines with structured XML output and metadata for multi-stage workflows.
- Success criteria: Prompt folder created in .claude/workspace/prompts/, SUMMARY.md generated, chain provenance maintained.
- It provides a Phase-based workflow: Intake, Chain Detection, Generate, Present, Execute, Summarize.
- It includes a folder structure example showing how prompts and summaries are organized under .claude/workspace/prompts/ with numbered subfolders and SUMMARY.md in each.
- It offers a Pattern-driven approach: Autonomous Purpose Detection to infer intent, followed by chain detection, purpose-specific questions, and a decision gate before generating prompts.
- It outlines multiple patterns (Workflow, Autonomous Purpose Detection, Purpose Templates) and detailed XML/plan/research/refine templates for structuring outputs.
- It references a single declared tool: claude-code.
When to use it
- When creating optimized prompts for complex workflows, delegating to subagents, or managing multi-stage execution.
- Not for simple single-turn prompts, manual messaging, or non-Claude integrations.
What it can touch
- It specifies a workspace path: .claude/workspace/prompts/ for storing prompts and SUMMARY.md.
- It mentions file outputs like SUMMARY.md and various XML/markdown structured artifacts in the workflow process.
Caveats
- License: MIT.
- Declared tool: claude-code.
- The skill is focused on multi-stage, Claude-to-Claude scenarios and explicitly excludes simple one-turn prompts and non-Claude integrations.
<mission_control> <objective>Generate meta-prompts for Claude-to-Claude pipelines with structured XML output and metadata for multi-stage workflows.</objective> <success_criteria>Prompt folder created in .claude/workspace/prompts/, SUMMARY.md generated, chain provenance maintained</success_criteria> </mission_control> ## Quick Start **If you need to create a single prompt:** Follow Phase 1 (Intake) → Phase 2 (Generate) → Phase 3 (Present). **If you need a prompt chain:** Use Chain Detection → Generate multiple prompts → Execute sequentially with dependency handling. **If you need purpose-specific templates:** Reference ## PATTERN: Purpose Templates for Do/Plan/Research/Refine patterns. ## Navigation | If you need... | Read this section... | | :------------- | :------------------- | | Create a meta-prompt | ## PATTERN: Workflow | | Purpose-specific templates | ## PATTERN: Purpose Templates | | Purpose inference | ## PATTERN: Autonomous Purpose Detection | | Chain detection and dependencies | ## PATTERN: Dependency Management | | Execution modes | ## PATTERN: Execution Engine | | SUMMARY.md structure | ## PATTERN: SUMMARY.md Template | | Common mistakes | ## ANTI-PATTERN: Common Mist
- Quick Start
- Navigation
- PATTERN: Workflow
- Folder Structure
- PATTERN: Autonomous Purpose Detection
- 1. Purpose Gate (CRITICAL - First Action)
- 2. Topic Identifier
- 3. Chain Detection
- 4. Purpose-Specific Questions
- 5. Decision Gate
- PATTERN: Purpose Templates
- Do Pattern (Execution)
- Plan Pattern (Roadmaps)
- Summary
What does the create-meta-prompts skill do?
Generate meta-prompts for Claude-to-Claude pipelines and multi-stage workflows. Use when creating optimized prompts for complex workflows, delegating to subagents, or managing multi-stage execution. Includes XML output structuring, metadata injection, and chain provenance. Not for simple single-turn prompts, manual messaging, or non-Claude integrations. Keywords: prompt chain, delegation, multi-stage, workflow orchestration, Claude-to-Claude.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill create-meta-prompts-git-fg-meta-plugin-manager --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.