Product Discovery Skills in pm-product-discovery: The Complete 13-Skill Toolkit

The pm-product-discovery package provides 13 reusable, markdown-based product discovery skills that enable product managers to execute the full discovery lifecycle programmatically, from ideation and assumption mapping to customer interviews and feature prioritization.

The phuryn/pm-skills open-source repository structures product management workflows as callable, version-controlled skills. The pm-product-discovery module specifically implements these product discovery skills as self-contained markdown files with corresponding CLI/Chat-Ops command façades, allowing teams to standardize discovery processes across the command line or integrated chat platforms.

The 13 Core Product Discovery Skills

The pm-product-discovery module organizes capabilities into 13 distinct skills, each residing as a SKILL.md file under pm-product-discovery/skills/<skill-name>/. These components cover four critical phases of product discovery: ideation, assumption validation, customer research, and strategic planning.

Ideation and Experimentation Skills

These four skills support divergent thinking and lean experimentation for both new and existing products:

• brainstorm-ideas-new: Generates fresh feature ideas for new products using a multi-perspective approach (PM, Designer, Engineer). Defined in pm-product-discovery/skills/brainstorm-ideas-new/SKILL.md.
• brainstorm-ideas-existing: Multi-perspective ideation specifically for enhancing products already in market. Defined in pm-product-discovery/skills/brainstorm-ideas-existing/SKILL.md.
• brainstorm-experiments-new: Creates lean-startup "pretotype" experiments to validate brand-new product ideas. Defined in pm-product-discovery/skills/brainstorm-experiments-new/SKILL.md.
• brainstorm-experiments-existing: Designs experiments to test assumptions for already-launched products. Defined in pm-product-discovery/skills/brainstorm-experiments-existing/SKILL.md.

Assumption Management Skills

These three skills help surface and rank risky assumptions before committing engineering resources:

• identify-assumptions-new: Surfaces risky assumptions across eight categories—including Go-to-Market and Team risks—for brand-new products. Source: pm-product-discovery/skills/identify-assumptions-new/SKILL.md.
• identify-assumptions-existing: Surfaces risky assumptions for existing products across four dimensions: Value, Usability, Viability, and Feasibility. Source: pm-product-discovery/skills/identify-assumptions-existing/SKILL.md.
• prioritize-assumptions: Ranks assumptions on an Impact × Risk matrix and suggests corresponding experiments. Source: pm-product-discovery/skills/prioritize-assumptions/SKILL.md.

Customer Research Skills

Structured interview capabilities for Jobs-to-be-Done (JTBD) research:

• interview-script: Builds a structured customer interview script including JTBD probing, warm-up, core questions, and wrap-up sections. Source: pm-product-discovery/skills/interview-script/SKILL.md.
• summarize-interview: Converts raw interview transcripts into structured templates identifying JTBD, satisfaction signals, and action items. Source: pm-product-discovery/skills/summarize-interview/SKILL.md.

Strategic Planning and Feature Management

Skills for mapping opportunities and managing the backlog:

• opportunity-solution-tree: Maps desired outcomes to opportunities, solutions, and experiments using the Opportunity Solution Tree (OST) framework. Source: pm-product-discovery/skills/opportunity-solution-tree/SKILL.md.
• analyze-feature-requests: Categorizes, themes, and ranks feature requests by impact, effort, risk, and strategic alignment. Source: pm-product-discovery/skills/analyze-feature-requests/SKILL.md.
• prioritize-features: Ranks backlog items by impact, effort, risk, and strategic alignment, delivering top-5 recommendations. Source: pm-product-discovery/skills/prioritize-features/SKILL.md.
• metrics-dashboard: Defines product metrics dashboards including key metrics, data sources, visualizations, and alerts. Source: pm-product-discovery/skills/metrics-dashboard/SKILL.md.

Invoking Product Discovery Skills via CLI

Each skill maps to a command through the repository's CLI/Chat-Ops façade. Commands follow the pattern pm-product-discovery:<command> and process arguments to generate stakeholder-ready markdown reports.

The following examples demonstrate how to invoke these product discovery skills from the command line:

# Run a full discovery session for a new product

pm-product-discovery:discover \
  --product "Smart Home Hub" \
  --stage new

# Prioritize a feature backlog and return top-5 recommendations

pm-product-discovery:prioritize-features \
  --input features.csv \
  --objective "Increase monthly active users by 20%"

# Generate a JTBD interview script

pm-product-discovery:interview \
  --script \
  --product "AI-Powered Calendar" \
  --jtbd "Schedule meetings quickly"

# Summarize an interview transcript into structured insights

pm-product-discovery:interview \
  --summarize \
  --transcript interview.txt

# Build an Opportunity-Solution Tree

pm-product-discovery:opportunity-solution-tree \
  --problems problems.txt \
  --output ost.md

Each command loads the corresponding SKILL.md template from pm-product-discovery/skills/<skill-name>/, processes the provided arguments, and outputs a markdown report.

File Structure and Architecture

The pm-product-discovery package separates knowledge from execution through a strict directory convention:

• Skill definitions: pm-product-discovery/skills/<skill>/SKILL.md contains the detailed prompt templates, context variables, and step-by-step instructions for each skill.
• Command wrappers: pm-product-discovery/commands/<command>.md maps CLI/Chat-Ops commands to underlying skills with human-readable documentation.
• Package documentation: pm-product-discovery/README.md provides the overview table of all 13 skills and quick-reference links.
• Repository documentation: Root README.md describes the broader pm-skills suite and links to each product area.

Summary

• The pm-product-discovery module contains 13 modular product discovery skills stored as markdown files in pm-product-discovery/skills/.
• Skills cover the full discovery lifecycle: ideation (brainstorm-ideas-*), assumption management (identify-assumptions-*, prioritize-assumptions), customer research (interview-script, summarize-interview), and strategic planning (opportunity-solution-tree, prioritize-features, metrics-dashboard).
• Each skill is invoked via CLI commands like pm-product-discovery:discover or pm-product-discovery:prioritize-features, which load the corresponding SKILL.md templates.
• The architecture separates knowledge (markdown skill files) from execution (command wrappers), enabling both programmatic invocation and manual reference.

Frequently Asked Questions

What is the difference between the "new" and "existing" product skills?

The "new" product skills target pre-launch or ideation-stage products with broader risk categories (8 types including Go-to-Market and Team), while "existing" product skills focus on post-launch optimization using the four standard product risks: Value, Usability, Viability, and Feasibility. For example, identify-assumptions-new evaluates entirely new market entries, whereas identify-assumptions-existing validates feature additions to current products.

How does the prioritize-assumptions skill work?

The prioritize-assumptions skill, defined in pm-product-discovery/skills/prioritize-assumptions/SKILL.md, takes a list of identified assumptions and maps them onto an Impact × Risk matrix. It automatically ranks assumptions by their potential to derail the product if wrong and suggests specific experiments to de-risk the top-ranked items, outputting a prioritized action plan.

Can I customize the interview-script template for non-JTBD interviews?

Yes. While the default interview-script skill in pm-product-discovery/skills/interview-script/SKILL.md is optimized for Jobs-to-be-Done (JTBD) research with specific warm-up and probing sections, the markdown-based structure allows modification of the underlying SKILL.md file to support other research methodologies like solution interviews or usability testing protocols.

Where are the skill commands defined in the repository?

Command mappings are stored in pm-product-discovery/commands/<command>.md, while the actual skill logic resides in pm-product-discovery/skills/<skill-name>/SKILL.md. The repository's CLI façade bridges these layers, parsing arguments like --product or --transcript and injecting them into the skill templates before execution.

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