Claude Code Plugins

Community-maintained marketplace

Feedback

Create/Updates detailed user personas and prioritized user stories based on target users and market research. Use when user asks to create personas, define user stories, or start Stage 2 after market research or init-ideas.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name dev-swarm-personas
description Create/Updates detailed user personas and prioritized user stories based on target users and market research. Use when user asks to create personas, define user stories, or start Stage 2 after market research or init-ideas.

AI Builder - Personas & User Stories

This skill creates/updates detailed user personas and prioritized user stories to define who will use the product and what they need to accomplish.

When to Use This Skill

  • User asks to "create personas" or "define user stories"
  • User requests to start Stage 2 or the next stage after market-research
  • User wants to define target users in detail
  • User wants to create/update user stories

Prerequisites

This skill requires 00-init-ideas to be completed. Market research (01-market-research) is optional.

Your Roles in This Skill

  • Product Manager: Create user stories in "As a [role], I want [feature] so that [benefit]" format. Prioritize features based on user needs and business goals (P0/P1/P2). Define acceptance criteria for each feature.
  • UX Designer: Create user personas based on target user research. Map user journeys and identify user needs and pain points. Ensure personas reflect real user behaviors and goals.

Role Communication

As an expert in your assigned roles, you must announce your actions before performing them using the following format:

As a {Role} [and {Role}, ...], I will {action description}

This communication pattern ensures transparency and allows for human-in-the-loop oversight at key decision points.

Instructions

Follow these steps in order:

Step 0: Verify Prerequisites and Gather Context

  1. Check if 00-init-ideas/ exists (mandatory):

    • If NOT found: Inform user they need to init ideas first, then STOP
    • If found: Read all files to understand:
      • Problem statement
      • Target users
      • Value proposition
      • Owner requirements
      • Cost budget (to understand constraints for this stage)
  2. Check if 01-market-research/ exists (optional):

    • If found: Read files to understand:
      • Target audience segments
      • Competitor user bases
      • Market validation findings
      • Gap analysis
    • If NOT found: Continue with just init-ideas data (acceptable for L2 projects)
  3. Check if this stage should be skipped:

    • Check if 02-personas/SKIP.md exists
    • If SKIP.md exists:
      • Read SKIP.md to understand why this stage was skipped
      • Inform the user: "Stage 2 (personas) is marked as SKIP because [reason from SKIP.md]"
      • Ask the user: "Would you like to proceed to the next stage (mvp)?"
      • If user says yes:
        • Exit this skill and inform them to run the next stage skill
      • If user says no:
        • Ask if they want to proceed with personas anyway
        • If yes, delete SKIP.md and continue with this skill
        • If no, exit the skill
  4. Check if 02-personas/ folder exists:

    • If exists: Read all existing files to understand current state
    • If NOT exists: Will create new structure
  5. If README.md exists: Check whether it requires diagrams. If it does, follow dev-swarm/docs/mermaid-diagram-guide.md and use the dev-swarm-mermaid skill to render outputs.

  6. Proceed to Step 1 with gathered context

Step 1: Refine Design Requirements in README and Get Approval

CRITICAL: Create/update README.md first based on previous stage results, get user approval, then create other docs.

  1. Analyze information from previous stages:

    • Read 00-init-ideas/ to understand problem, target users, value proposition
    • Read 01-market-research/ (if exists) to understand market segments and user validation
    • Consider cost-budget constraints for this stage
  2. Create or update 02-personas/README.md with refined requirements:

    • List deliverables explicitly in README (typical: persona-primary.md, persona-secondary.md, user-stories.md)
    • Stage overview and objectives (based on previous stage context)
    • Owners: Product Manager, UX Designer
    • Diagrams (if required by project init):
      • Reference dev-swarm/docs/mermaid-diagram-guide.md
      • Include diagram/ deliverables when needed
    • What personas will be created:
      • How many personas (1-2 typically)
      • What market research insights will inform personas
      • What user stories will be created (P0/P1/P2)
    • Methodology:
      • How personas will be created (based on init-ideas + market research)
      • User story format and prioritization approach
    • Deliverables planned:
      • List of files that will be created (persona-primary.md, user-stories.md, etc.)
    • Budget allocation for this stage (from cost-budget.md)
    • Status: In Progress (update to "Completed" after implementation)
  3. Present README to user:

    • Show the personas approach and what will be created
    • Show what documentation files will be created
    • Explain how it aligns with previous stages
    • Ask: "Does this personas plan look good? Should I proceed with creating personas and user stories?"
  4. Wait for user approval:

    • If user says yes: Proceed to Step 2
    • If user says no:
      • Ask what needs to be changed
      • Update README based on feedback
      • Ask for approval again

Step 2: Create/Update Personas Structure

Only after user approves the README:

  1. Create files as specified in the approved README.md:

    IMPORTANT: The file structure below is a SAMPLE only. The actual files you create must follow what was approved in the README.md in Step 1.

    Typical structure (example):

    02-personas/
    ├── README.md (already created and approved in Step 1)
    ├── persona-primary.md (if specified in README)
    ├── persona-secondary.md (if specified in README)
    └── user-stories.md (if specified in README)
    

    Create only the files listed in the README's "Deliverables planned" section.

Step 3: Create/Update User Personas

IMPORTANT: Only create personas after README is approved in Step 1.

NOTE: The content structure below provides GUIDELINES for typical persona content. Adapt based on the approved README and project needs.

persona-primary.md (if specified in README):

Create a detailed primary persona including:

  • Basic Information:

    • Name (fictional but relatable)
    • Age range
    • Occupation/Role
    • Location/Context
    • Photo or avatar description (optional)
  • Background:

    • Professional background
    • Technical proficiency level
    • Relevant experience
  • Goals & Motivations:

    • Primary goals when using this product
    • What success looks like for them
    • Key motivations and drivers
  • Pain Points & Frustrations:

    • Current challenges and problems
    • Frustrations with existing solutions
    • Unmet needs
  • Behaviors & Preferences:

    • How they currently solve the problem
    • Preferred tools and platforms
    • Usage patterns and habits
    • Communication preferences
  • Needs from This Product:

    • Must-have features (P0)
    • Important features (P1)
    • Nice-to-have features (P2)
  • Quote:

    • A fictional quote that captures their mindset

persona-secondary.md (optional):

If there's a distinct secondary user segment, create a second persona following the same structure. Only create this if:

  • There's a clearly different user segment
  • Their needs differ significantly from primary persona
  • They represent a meaningful portion of target users

If secondary persona is not needed, you can skip creating this file.

Step 4: Create/Update User Stories

NOTE: The content structure below provides GUIDELINES for typical user stories. Adapt based on the approved README and project needs.

user-stories.md (if specified in README):

Create prioritized user stories using the format: "As a [role], I want [feature] so that [benefit]"

Organize stories by priority:

P0 - Must Have (Core Features):

  • Critical features that deliver the core value proposition
  • Without these, the product doesn't solve the problem
  • 5-10 user stories typically

P1 - Should Have (Important Features):

  • Important features that enhance the experience
  • Significantly improve usability or value
  • 5-15 user stories typically

P2 - Nice to Have (Enhancement Features):

  • Features that add polish or convenience
  • Not critical for initial launch
  • Can be deferred to later versions
  • 5-10 user stories typically

Format for each user story:

### [Priority] - [Story Title]

**User Story:**
As a [persona name/role],
I want [specific capability],
So that [benefit/value achieved].

**Acceptance Criteria:**
- [ ] Criterion 1
- [ ] Criterion 2
- [ ] Criterion 3

**Notes:**
- Any additional context
- Related stories or dependencies
- Technical considerations (if any)

Step 5: Ensure Traceability

Make sure user stories map back to:

  • Problem statement from 00-init-ideas
  • Value proposition from 00-init-ideas
  • Gap analysis from 01-market-research (if available)
  • Owner requirements from 00-init-ideas

Step 6: Final User Review

  1. Inform user that personas and user stories are complete

  2. Update README.md:

    • Change Status from "In Progress" to "Completed"
    • Add a Summary section with key insights (2-3 paragraphs)
    • Add a Created Files section listing all created files
  3. Present completed work to user:

    • Number of personas created
    • P0/P1/P2 story distribution
    • How stories address the core problem
  4. Ask if they want to proceed to the next stage (MVP definition)

  5. Make adjustments based on user feedback if needed

Step 7: Commit to Git (if user confirms)

  1. If user confirms personas and user stories are complete:
    • Ask if they want to commit to git
  2. If user wants to commit:
    • Stage all changes in 02-personas/
    • Commit with message: "Define user personas and prioritized user stories (Stage 2)"

Expected Project Structure

project-root/
├── 00-init-ideas/
│   └── [existing files]
├── 01-market-research/ (optional)
│   └── [existing files if present]
└── 02-personas/
    ├── README.md (with owners and summary)
    ├── persona-primary.md
    ├── persona-secondary.md (optional)
    └── user-stories.md (P0/P1/P2 prioritized)

Key Principles

  • Create realistic, relatable personas based on actual target users
  • Focus on goals, pain points, and behaviors, not just demographics
  • Write user stories from the user's perspective, not the business perspective
  • Prioritize ruthlessly - P0 should be minimal core features only
  • Use clear acceptance criteria that can be tested
  • Ensure every story delivers user value
  • Trace stories back to problem statement and value proposition
  • Keep stories small and specific enough to implement

User Story Best Practices

  1. Focus on user value: Every story should deliver tangible benefit
  2. Keep stories independent: Each story should be implementable separately
  3. Make stories testable: Acceptance criteria should be verifiable
  4. Use persona names: Reference specific personas in stories
  5. Avoid technical details: Focus on "what" and "why", not "how"
  6. Prioritize based on value: P0 = MVP, P1 = Important, P2 = Nice-to-have

Deliverables

By the end of this stage, you should have:

  • 1-2 detailed user personas representing target users
  • 15-35 user stories organized by priority (P0/P1/P2)
  • Clear acceptance criteria for each story
  • Traceability from stories to problem statement and value proposition
  • Foundation for MVP scope definition