Understanding the PM Skills Command: Orchestrating Product Management Workflows

A PM Skills Command is a user-triggered workflow that strings together one or more Skills to deliver a complete, end-to-end product-management task through simple slash commands.

The phuryn/pm-skills repository provides a modular framework where high-level PM Skills Commands serve as the primary interface between product managers and complex AI-driven workflows. These commands abstract underlying implementation details while exposing powerful capabilities through intuitive syntax documented in dedicated markdown files.

Core Purpose and Architecture

A PM Skills Command functions as the human-facing entry point to the system's capabilities. When you execute a slash command such as /discover or /review-resume, the system automatically loads the underlying Skills, runs them in sequence, and returns structured output.

This architecture delivers three key benefits:

• Abstraction: Commands hide individual skill implementations, presenting a single, easy-to-use action rather than exposing complex orchestration logic to end users.
• Reusability: The same skill definitions can be reused across multiple commands, allowing the review-resume skill to power both standalone reviews and tailored resume adjustments.
• Documentation: Each command lives in its own markdown file (e.g., pm-toolkit/commands/review-resume.md) and includes exact syntax and usage examples.

According to the repository README, commands represent the mechanism by which AI assistants (Claude, Codex, Gemini, etc.) expose high-level capabilities while maintaining clean separation from skill definitions.

Command Syntax and Usage Examples

Commands follow a consistent slash-command pattern that accepts arguments and attachments. The exact syntax varies by workflow complexity, from simple single-skill invocations to multi-step chains.

Resume Review and Tailoring Workflows

/review-resume [attach your PM resume]

This command runs the review-resume skill, which evaluates the resume against 10 best-practice criteria defined in pm-toolkit/commands/review-resume.md.

/tailor-resume [attach resume] [paste job description]

This invocation chains the review-resume skill with a tailoring step to match a specific job posting, leveraging the implementation in pm-toolkit/commands/tailor-resume.md.

Strategy and Discovery Commands

/strategy B2B project-management tool for agencies

This command invokes the product-strategy skill (plus related sub-skills) to produce a full 9-section Product Strategy Canvas, as defined in pm-product-strategy/commands/strategy.md.

/discover

The discovery workflow entry point lives in pm-product-discovery/commands/discover.md and executes an end-to-end discovery process without requiring additional parameters.

Go-to-Market Execution

/plan-launch AI code-review tool targeting mid-size engineering teams

This executes the GTM workflow—covering beachhead identification, ICP definition, messaging, and launch plan generation—implemented in pm-go-to-market/commands/plan-launch.md.

Command File Organization

The repository organizes commands by functional domain, with each command existing as a standalone markdown file containing documentation and metadata:

Plugin	Command File	Function
pm-toolkit	pm-toolkit/commands/review-resume.md	Full PM resume review
pm-toolkit	pm-toolkit/commands/tailor-resume.md	Resume tailoring to job descriptions
pm-product-discovery	pm-product-discovery/commands/discover.md	End-to-end discovery workflow
pm-product-strategy	pm-product-strategy/commands/strategy.md	Product Strategy Canvas generation
pm-go-to-market	pm-go-to-market/commands/plan-launch.md	GTM launch plan creation
pm-data-analytics	pm-data-analytics/commands/write-query.md	SQL generation from natural language

Each file path follows the pattern {plugin-name}/commands/{command-name}.md, ensuring discoverability and maintainability across the codebase.

Summary

• PM Skills Commands provide user-friendly entry points that orchestrate underlying skills into complete product-management outcomes.
• Commands are implemented as documented markdown files within plugin-specific commands/ directories, such as pm-toolkit/commands/review-resume.md and pm-go-to-market/commands/plan-launch.md.
• The slash-command interface (/review-resume, /strategy, etc.) abstracts complex skill chains while enabling reuse across different workflows.
• This design allows AI assistants to expose high-level capabilities without duplicating skill logic, as implemented in the phuryn/pm-skills repository.

Frequently Asked Questions

What is the difference between a PM Skills Command and a Skill?

A Skill represents a discrete capability or atomic task (such as evaluating a resume against specific criteria), while a PM Skills Command is the user-facing workflow that orchestrates one or more skills to deliver an end-to-end result. Commands hide the underlying skill implementations and manage execution sequencing, allowing the same skill to be reused across multiple command contexts according to the repository architecture.

Where are command definitions stored in the repository?

Each command definition resides in its own markdown file within the commands/ subdirectory of its respective plugin. For example, the /review-resume command is documented in pm-toolkit/commands/review-resume.md, while the /plan-launch command lives in pm-go-to-market/commands/plan-launch.md. This structure separates command documentation from skill implementation logic.

How do PM Skills Commands integrate with AI assistants?

Commands serve as the integration layer between users and AI assistants such as Claude, Codex, and Gemini. When a user types a slash command, the AI assistant loads the corresponding command definition from the appropriate commands/*.md file, executes the associated skills in sequence, and returns structured output. This design allows AI systems to expose high-level product management capabilities while reusing standardized skill definitions across different command workflows.

Can a single command invoke multiple skills?

Yes. Commands frequently chain multiple skills to accomplish complex tasks. For instance, the /tailor-resume command combines the review-resume skill with additional tailoring logic to adapt content for specific job descriptions. Similarly, the /strategy command coordinates the product-strategy skill with related sub-skills to generate comprehensive 9-section Product Strategy Canvases.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}