Complete Discovery Cycle Facilitated by the `/discover` Command: A 7-Step Technical Guide
The /discover command in the phuryn/pm-skills repository executes a structured seven-step discovery workflow that progresses from initial context gathering to experiment design, utilizing specialized skill modules to validate product assumptions before engineering commitment.
The complete discovery cycle provides product managers with a repeatable, checkpointed framework for rapidly validating ideas. Implemented as a slash-command in the open-source phuryn/pm-skills repository, this orchestration layer chains together discrete skill modules to move from vague concepts to concrete experiment plans within 15-30 minutes.
The Seven Steps of the Complete Discovery Cycle
The /discover command defined in pm-product-discovery/commands/discover.md sequences seven distinct phases, each implemented by dedicated skill files that handle specific discovery concerns.
Step 1: Context Gathering
The cycle initiates within [discover.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/commands/discover.md) by capturing foundational inputs. The interface prompts for the specific product or feature under investigation, existing known information, and the critical business decision the discovery process will ultimately inform.
Step 2: Multi-Perspective Ideation
Based on product maturity, the workflow invokes either [brainstorm-ideas-existing/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/brainstorm-ideas-existing/SKILL.md) or [brainstorm-ideas-new/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/brainstorm-ideas-new/SKILL.md). These modules generate cross-functional ideas from Product Management, Design, and Engineering perspectives, presenting candidates for user selection before proceeding.
Step 3: Assumption Mapping
Selected ideas undergo devil's-advocate analysis through [identify-assumptions-existing/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/identify-assumptions-existing/SKILL.md) or [identify-assumptions-new/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/identify-assumptions-new/SKILL.md). This step surfaces risky assumptions across Value, Usability, Feasibility, and Viability dimensions, with the "new" variant adding Go-to-Market analysis for zero-to-one products.
Step 4: Risk-Based Prioritization
Using [prioritize-assumptions/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/prioritize-assumptions/SKILL.md), the system maps each assumption onto an Impact × Risk matrix. This identifies "leap-of-faith" assumptions and groups testable hypotheses according to validation priority and business risk.
Step 5: Experiment Design
The workflow transitions to validation planning via [brainstorm-experiments-existing/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/brainstorm-experiments-existing/SKILL.md) or [brainstorm-experiments-new/SKILL.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/skills/brainstorm-experiments-new/SKILL.md). Each high-priority assumption receives 1-2 proposed low-effort validation methods, such as A/B tests, fake-door landing pages, pretotypes, or concierge MVPs.
Step 6: Discovery Plan Compilation
An inline template within [discover.md](https://github.com/phuryn/pm-skills/blob/main/pm-product-discovery/commands/discover.md) synthesizes the selected ideas, prioritized assumptions, and chosen experiments into a downloadable markdown discovery plan document that serves as the single source of truth.
Step 7: Next-Step Recommendations
Finally, the command presents contextual follow-up actions including PRD creation, interview script drafting, metrics dashboard setup, or engineering effort estimation, defined inline within the command file to close the loop.
Checkpointed Execution Architecture
Each step in the complete discovery cycle is checkpointed: the assistant pauses for user confirmation or redirection between phases. This architecture, implemented in the main command file, allows the workflow to be extended, deep-dived, or pivoted without losing accumulated context. The skill modules communicate via structured outputs that feed sequentially into subsequent stages, ensuring data integrity across the entire pipeline.
Practical Usage Examples
Trigger the complete discovery cycle with a single slash command:
/discover Smart notification system for our project-management tool
The assistant executes the seven-step sequence:
Step 1 – What are you exploring? … (context questions)
Step 2 – Here are 10 ideas (PM/Designer/Engineer). Which 3-5 should we test?
Step 3 – Identified assumptions for each selected idea …
Step 4 – Prioritized assumptions (Impact × Risk matrix) …
Step 5 – Proposed experiments: fake-door landing page, A/B test of notification toggle, …
Step 6 – Generating a markdown Discovery Plan (downloadable file) …
Step 7 – Would you like a PRD, interview script, or metrics dashboard next?
For an interactive session with explicit checkpoint control:
/discover
The assistant asks: "What product or feature are you discovering?"
User replies: "AI-powered meeting summarizer for remote teams."
Assistant proceeds through the seven steps, prompting for selections after each stage.
Key Source Files
Summary
- The complete discovery cycle consists of seven sequential steps: Context Gathering, Multi-Perspective Ideation, Assumption Mapping, Risk-Based Prioritization, Experiment Design, Discovery Plan Compilation, and Next-Step Recommendations.
- The workflow is checkpointed between each step, allowing user redirection while maintaining process integrity.
- Execution time is optimized for 15-30 minutes through efficient skill chaining in
pm-product-discovery/commands/discover.md. - Parallel skill sets exist for existing versus new products, differing primarily in Go-to-Market analysis and validation depth.
Frequently Asked Questions
How long does the complete discovery cycle take?
According to the implementation in pm-product-discovery/commands/discover.md, the entire workflow is designed to complete in 15-30 minutes. This rapid turnaround is achieved through checkpointed interactions where the user provides targeted inputs at each step, preventing scope creep while maintaining thoroughness.
What is the difference between the "existing" and "new" product skill paths?
The command automatically selects between parallel skill sets based on product maturity. For existing products, it invokes brainstorm-ideas-existing, identify-assumptions-existing, and brainstorm-experiments-existing, which focus on feature enhancements and incremental validation; for new products, the "new" variants add Go-to-Market assumption analysis and emphasize zero-to-one validation strategies.
Can I skip steps in the discovery cycle?
No, the workflow enforces sequential completion because each step's output feeds directly into the next stage's input. However, the checkpointed architecture allows you to redirect or deep-dive at any pause point, effectively customizing the depth of each stage without breaking the dependency chain.
Where is the discovery plan template defined?
The final discovery plan compilation logic resides inline within the discover.md command file, not in a separate skill module. This template aggregates outputs from the prior six steps into a structured markdown document that serves as the single source of truth for your validation strategy.
Have a question about this repo?
These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:
curl -s "https://instagit.com/install.md" 
Maintain an open-source project? Get it listed too →