meta-command-creator

name	meta-command-creator
description	Create Pi prompt templates (slash commands) for repetitive tasks. Use when adding /commands that trigger specific prompts or workflows.
license	MIT
compatibility	pi
metadata	{"audience":"developers","category":"meta"}

Meta Command Creator

Create prompt templates (slash commands) that trigger specific prompts in Pi.

What I Do

• Generate prompt template files (Markdown with YAML frontmatter)
• Configure command arguments and placeholders
• Set up skill injection and model overrides
• Help design reusable workflow commands

When to Use Me

Use this skill when:

• Creating reusable prompt workflows (e.g., /test, /review, /deploy)
• Building project-specific commands
• Defining commands that load specific skills
• Adding shortcuts for common tasks

Do NOT use for:

• Creating tools the LLM calls (use meta-tool-creator)
• Creating extensions (use meta-extension-creator)
• Creating agents (use meta-agent-creator)
• Creating skills (use meta-skill-creator)

Quick Start

Basic Prompt Template

Create .pi/prompts/test.md:

---
description: Run tests with coverage
---

Run the full test suite with coverage report and show any failures.
Focus on the failing tests and suggest fixes.

Run it with:

/test

Command with Arguments

Create .pi/prompts/component.md:

---
description: Create a new component
---

Create a new React component named $ARGUMENTS with TypeScript support.
Include proper typing and basic structure.

Run with arguments:

/component Button

Positional Arguments

Use $1, $2, $3 for specific arguments:

---
description: Create a file with content
---

Create a file named $1 in the directory $2
with the following content: $3

Run:

/create-file config.json src "{ \"key\": \"value\" }"

Prompt Template Location

Location	Scope	Path
Project	This project only	.pi/prompts/
Global	All projects	~/.pi/agent/prompts/

The filename becomes the command name:

• test.md → /test
• review-pr.md → /review-pr

Frontmatter Options

---
description: Brief description shown in command list
skill: git, github          # Skills to load with this prompt
model: ${models.medium}  # Override model for this command
---

Your prompt template here...

Option	Type	Required	Description
description	string	Yes	Brief description shown in / menu
skill	string	No	Comma-separated skills to inject
model	string	No	Override the default model

Prompt Features

Shell Output Injection

Use `command` in the prompt body:

---
description: Analyze test coverage
---

Here are the current test results:
`npm test`

Based on these results, suggest improvements.

File References

Reference files for context:

---
description: Review component
---

Review the component. Check for performance issues.
Load @src/components/Button.tsx for context.

Common Patterns

Testing Commands

---
description: Run and fix failing tests
skill: git
---

Run the test suite and analyze failures.
Fix any issues found.

Code Review Commands

---
description: Review changes
skill: git, github
---

Review the current changes:
`git diff HEAD~1`

Check for:
- Code quality issues
- Security concerns
- Performance problems

Documentation Commands

---
description: Generate docs for file
---

Generate documentation for $ARGUMENTS.
Follow project conventions.

Handoff Command

---
description: Create handoff for session continuation
skill: context-management
---

Create a handoff document for the current session.
Focus: $ARGUMENTS

1. Summarize what was accomplished
2. Document blockers and next steps
3. Save to workspace/users/$USER/handoffs/

Feedback Command

---
description: Submit feedback or report issues
skill: github
---

The user wants to provide feedback: $ARGUMENTS

Help them:
1. Clarify the type (bug, feature, feedback)
2. Gather relevant details
3. Create a GitLab issue with context

Command vs Tool vs Skill

Feature	Command	Tool	Skill
Purpose	User-triggered prompts	LLM-callable functions	Knowledge/instructions
Invoked by	User types /name	LLM decides to call	Agent loads on-demand
Location	.pi/prompts/	.pi/extensions/	.pi/skills/
Format	Markdown + frontmatter	TypeScript	Markdown + frontmatter
Arguments	$ARGUMENTS, $1, $2	TypeBox schema	N/A

Tips

1. Keep prompts focused: One command, one purpose
2. Use skills: Load relevant skills via frontmatter
3. Inject context: Use shell commands for dynamic data
4. Document well: Clear descriptions help discoverability
5. Test arguments: Verify $ARGUMENTS and $1 work as expected