Lean Canvas Skill for Claude
Document your business model on one page and systematically de-risk it. Master Ash Maurya's adaptation of Business Model Canvas optimized for startups and uncertainty. Use when: **Starting a new venture** to articulate and test your business model; **Preparing for customer discovery** to document hypotheses to validate; **Pivoting decisions** to compare alternative business models; **Investor conversations** to communicate your model concisely; **Team alignment** to get everyone on the same page
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/guia-matthieu/clawfu-skills --skill lean-canvasThe instruction itself
28 sections, as written by the authorLean Canvas
> Document your business model on one page and systematically de-risk it. Master Ash Maurya's adaptation of Business Model Canvas optimized for startups and uncertainty.
When to Use This Skill
- Starting a new venture to articulate and test your business model
- Preparing for customer discovery to document hypotheses to validate
- Pivoting decisions to compare alternative business models
- Investor conversations to communicate your model concisely
- Team alignment to get everyone on the same page
- Comparing opportunities to evaluate multiple ideas systematically
Methodology Foundation
| Aspect | Details |
|--------|---------|
| Source | Ash Maurya - "Running Lean" (2012), adapted from Osterwalder's Business Model Canvas |
| Core Principle | "Document your Plan A, identify the riskiest parts, and systematically test them." |
| Why This Matters | A business plan is a 60-page guess. A Lean Canvas is a 1-page hypothesis you can test in weeks, not months. It replaces planning with learning. |
What Claude Does vs What You Decide
| Claude Does | You Decide |
|-------------|------------|
| Structures production workflow | Final creative direction |
| Suggests technical approaches | Equipment and tool choices |
| Creates templates and checklists | Quality standards |
| Identifies best practices | Brand/voice decisions |
| Generates script outlines | Final script approval |
What This Skill Does
- Creates one-page business models - 9 boxes that capture your entire model
- Identifies riskiest assumptions - Highlights what could kill your business
- Prioritizes validation experiments - Focuses on highest-risk unknowns first
- Enables rapid pivots - Easy to update as you learn
- Facilitates communication - Share your model in 5 minutes
- Tracks evolution - Version control your business model thinking
How to Use
Create a Lean Canvas for a New Idea
Create a Lean Canvas for this business idea: [description]
Fill out all 9 boxes and identify the top 3 riskiest assumptions.
Compare Two Business Models
I'm deciding between two approaches:
Option A: [description]
Option B: [description]
Create Lean Canvases for both and compare them on risk and potential.
Identify What to Validate First
Here's my Lean Canvas: [paste canvas]
What are the riskiest assumptions? Design experiments to test them.
Instructions
When creating or analyzing Lean Canvases, follow this systematic approach:
Step 1: Understand the 9 Boxes
## Lean Canvas Structure
┌──────────────────┬──────────────────┬──────────────────┐
│ │ │ │
│ 2. PROBLEM │ 4. SOLUTION │ 3. UNIQUE VALUE │
│ (Top 3) │ (Top 3 features)│ PROPOSITION │
│ │ │ │
│ │ │ High-level │
│ ├──────────────────┤ concept │
│ │ │ │
│ Existing │ 8. KEY METRICS │ │
│ Alternatives │ (Pirates: │ │
│ │ AARRR) │ │
│ │ │ │
├──────────────────┼──────────────────┼──────────────────┤
│ │ │ │
│ 9. UNFAIR │ 5. CHANNELS │ 1. CUSTOMER │
│ ADVANTAGE │ (Path to │ SEGMENTS │
│ (Can't be │ customers) │ (Target users) │
│ copied) │ │ │
│ │ │ Early Adopters │
│ │ │ │
├──────────────────┴──────────────────┴──────────────────┤
│ 7. COST STRUCTURE │ 6. REVENUE STREAMS │
│ (Fixed + Variable) │ (Pricing model) │
└──────────────────────────────┴─────────────────────────┘
Key Difference from Business Model Canvas:
- Replaces Partners/Resources/Activities with Problem/Solution/Key Metrics
- Adds Unfair Advantage
- Focuses on RISK and LEARNING, not operational planning
Step 2: Fill Out Each Box (In Order)
Recommended Order: Customer Segments → Problem → Unique Value Proposition → Solution → Channels → Revenue → Cost → Key Metrics → Unfair Advantage
## Box-by-Box Guide
### 1. CUSTOMER SEGMENTS
**Question:** Who are you creating value for?
Target customers:
- [Primary segment]
- [Secondary segment if any]
Early Adopters (most important):
- [Specific description of first customers]
- Why they'll buy first: [reason]
Tips:
- Be specific (not "businesses" but "SaaS companies 10-50 employees")
- Identify early adopters who feel the pain most acutely
- If you can't describe them, you can't find them
---
### 2. PROBLEM
**Question:** What problems are you solving?
Top 3 Problems:
1. [Most critical problem]
2. [Second problem]
3. [Third problem]
Existing Alternatives (how they solve it today):
- [Alternative 1]
- [Alternative 2]
Tips:
- List problems from the CUSTOMER's perspective
- If existing alternatives work well, your problem isn't painful enough
- Every problem should be something you've validated (or will validate first)
---
### 3. UNIQUE VALUE PROPOSITION
**Question:** Why should customers choose you?
Single clear message:
"[We help] [customer segment] [achieve outcome] [unlike alternatives]
[because unique differentiator]."
High-level concept (analogy):
"X for Y" or "Like X but for Y"
Example: "Uber for dog walkers"
Tips:
- Focus on the END BENEFIT, not features
- Make it different, not just better
- Test: Can you say this in 10 seconds?
---
### 4. SOLUTION
**Question:** What are you building?
Top 3 Features (that solve top 3 problems):
1. [Feature → Problem 1]
2. [Feature → Problem 2]
3. [Feature → Problem 3]
Tips:
- Match each solution to a problem
- Keep it minimal - MVP thinking
- This box should be the LAST one you fill with certainty
---
### 5. CHANNELS
**Question:** How will you reach customers?
Path to Customers:
- Awareness: [How they learn about you]
- Acquisition: [How they start using]
- Retention: [How they keep using]
Specific channels:
- [Channel 1: e.g., Content marketing]
- [Channel 2: e.g., Direct sales]
- [Channel 3: e.g., Partnerships]
Tips:
- Start with channels that don't scale (do things that don't scale)
- Match channels to where early adopters spend time
- Free channels first, paid channels when you have product-market fit
---
### 6. REVENUE STREAMS
**Question:** How will you make money?
Pricing Model:
- [ ] One-time purchase
- [ ] Subscription
- [ ] Freemium
- [ ] Transaction fee
- [ ] Advertising
- [ ] Other: ___________
Price Point:
- [Price]: [Justification]
Revenue Formula:
- [Customers] × [Price] × [Frequency] = [Revenue]
Tips:
- Price on value, not cost
- Test pricing early (it's a feature)
- If you can't charge, you don't have a business
---
### 7. COST STRUCTURE
**Question:** What are your costs?
Fixed Costs (monthly):
- [Cost 1]: $___
- [Cost 2]: $___
- Total Fixed: $___
Variable Costs (per customer):
- [Cost per customer]: $___
Customer Acquisition Cost (target):
- CAC: $___
Break-even:
- Need ___ customers at $___ to break even
Tips:
- Keep fixed costs minimal early
- Know your unit economics before scaling
- CAC must be < LTV (lifetime value)
---
### 8. KEY METRICS
**Question:** How will you measure success?
Pirate Metrics (AARRR):
- Acquisition: [How many sign up?]
- Activation: [How many have "aha" moment?]
- Retention: [How many come back?]
- Revenue: [How many pay?]
- Referral: [How many refer others?]
One Metric That Matters (right now):
- [Single metric]: [Target]
Tips:
- Focus on ONE metric at a time
- Vanity metrics (signups, page views) lie
- Measure behavior, not opinions
---
### 9. UNFAIR ADVANTAGE
**Question:** What makes you defensible?
Can't be easily copied or bought:
- [ ] Insider information
- [ ] Dream team
- [ ] Personal authority/brand
- [ ] Network effects
- [ ] Community
- [ ] Existing customers
- [ ] Proprietary tech/data
- [ ] SEO ranking
Your unfair advantage:
- [What is it?]
- [Why can't competitors copy it?]
Tips:
- Most startups don't have one at first (that's OK)
- It often emerges over time
- "Passion" and "first mover" are NOT unfair advantages
Step 3: Identify Riskiest Assumptions
## Risk Assessment
### Stage 1 Risks (Product Risk)
"Do I have a problem worth solving?"
| Assumption | Type | Evidence | Risk Level |
|------------|------|----------|------------|
| Problem exists and is painful | PROBLEM | | High/Med/Low |
| Customers are identifiable | CUSTOMER | | High/Med/Low |
| Current alternatives inadequate | PROBLEM | | High/Med/Low |
### Stage 2 Risks (Market Risk)
"Have I built something people want?"
| Assumption | Type | Evidence | Risk Level |
|------------|------|----------|------------|
| Solution solves the problem | SOLUTION | | High/Med/Low |
| Customers will pay [price] | REVENUE | | High/Med/Low |
| CAC is sustainable | COST | | High/Med/Low |
### Stage 3 Risks (Scale Risk)
"Can I build a viable business?"
| Assumption | Type | Evidence | Risk Level |
|------------|------|----------|------------|
| Channels work at scale | CHANNEL | | High/Med/Low |
| Unit economics work | COST/REV | | High/Med/Low |
| Defensibility exists | ADVANTAGE | | High/Med/Low |
### TOP 3 RISKIEST ASSUMPTIONS
1. [Highest risk - validate first]
2. [Second highest]
3. [Third highest]
Step 4: Design Validation Experiments
## Validation Plan
### For each risky assumption:
**Assumption:** [Statement]
**Risk if wrong:** [Consequence]
**Experiment:**
- Type: [Interview / Landing page / Prototype / Concierge]
- Target: [Who/How many]
- Timeline: [Duration]
**Success Criteria:**
- Validated if: [Specific metric]
- Invalidated if: [Specific metric]
**Next Action:**
- [ ] [Specific next step]
Examples
Example 1: B2B SaaS - Sales Enablement Tool
Input:
> "Create a Lean Canvas for a sales enablement tool that helps SDRs personalize outreach at scale."
Output:
## LEAN CANVAS: SalesAI - Personalized Outreach at Scale
┌─────────────────────────────────────────────────────────────────────┐
│ │
│ 2. PROBLEM │ 4. SOLUTION │
│ ──────────── │ ──────────── │
│ 1. SDRs spend 40% of time │ 1. AI-generated personalized │
│ researching prospects │ intros based on LinkedIn + │
│ 2. Generic outreach gets │ company data │
│ <2% response rates │ 2. One-click personalization │
│ 3. Good reps leave, burnout │ for 100+ contacts/day │
│ from repetitive work │ 3. Response rate dashboard │
│ │ with A/B testing │
│ Existing Alternatives: │ │
│ - Manual research (slow) ├─────────────────────────────────│
│ - Outreach.io (templates only) │ 8. KEY METRICS │
│ - ZoomInfo (data, no writing) │ ──────────────── │
│ │ - # messages personalized/day │
│ │ - Response rate improvement │
│ │ - Time saved per SDR │
│ │ - OMTM: Response rate lift % │
│ │ │
├──────────────────────────────────┴──────────────────────────────────┤
│ 3. UNIQUE VALUE PROPOSITION │
│ ──────────────────────────── │
│ "Write 100 personalized emails in the time it takes to write 5." │
│ │
│ High-level concept: "Jasper AI for sales outreach" │
│ │
├─────────────────────────────────────────────────────────────────────┤
│ 9. UNFAIR ADVANTAGE │ 5. CHANNELS │ 1. │
│ ──────────────────── │ ──────────── │ CUST │
│ - Training data from │ - LinkedIn content │ SEGS │
│ 1M+ successful emails │ - Sales podcasts ads │ ──── │
│ - Network of SDR community │ - Outbound (dogfooding) │ B2B │
│ (early adopters) │ - Integrations: │ SDRs │
│ - (Initially: None) │ Outreach, Salesloft │ at │
│ │ │ 50- │
│ │ │ 500 │
│ │ │ emp │
│ │ │ SaaS │
│ │ │ │
│ │ │ EA: │
│ │ │ SDR │
│ │ │ mgrs │
│ │ │ w/ │
│ │ │ 5+ │
│ │ │ reps │
├─────────────────────────────────┴───────────────────────────┴──────┤
│ 7. COST STRUCTURE │ 6. REVENUE STREAMS │
│ ────────────────── │ ──────────────────── │
│ Fixed: │ Model: Per-seat SaaS │
│ - Team (2 founders): $0 (sweat) │ Price: $99/user/month │
│ - Infrastructure: $500/mo │ Tiers: │
│ - Tools: $200/mo │ - Starter: $99 (1-5 seats) │
│ │ - Growth: $79 (6-20 seats) │
│ Variable: │ - Enterprise: Custom │
│ - AI costs: ~$0.10/user/day │ │
│ - CAC target: <$500 │ LTV target: $1,800 (18 mo) │
│ │ LTV:CAC ratio: 3.6:1 │
│ Break-even: 50 paying users │ │
└─────────────────────────────────────┴──────────────────────────────┘
Riskiest Assumptions:
| # | Assumption | Risk Level | Why |
|---|------------|------------|-----|
| 1 | AI can write emails that perform better than humans | HIGH | Core value prop depends on this |
| 2 | SDRs/managers will trust AI with prospect communication | HIGH | Cultural barrier could block adoption |
| 3 | $99/user is acceptable price point | MEDIUM | Determines if business is viable |
Validation Plan:
- Week 1-2: Build prototype, A/B test AI vs. human emails (10 SDRs)
- Week 3-4: Interview 15 SDR managers on trust/adoption concerns
- Week 5: Pricing validation with early adopter interviews
Example 2: Consumer App - Fitness for Parents
Input:
> "Here's my Lean Canvas for a fitness app for busy parents. What are the riskiest assumptions?"
User's Canvas:
Problem: No time to exercise, gym is inconvenient
Customer: Working parents 30-45 with young kids
Solution: 10-minute home workouts, kid-friendly
Revenue: $9.99/month subscription
Output:
Risk Analysis:
CRITICAL ISSUE: This canvas is incomplete and under-specified.
Problems I See:
- Customer segment too broad
- "Working parents 30-45" = millions of people
- Who are the EARLY ADOPTERS who will pay first?
- Missing: Income level, fitness history, tech savviness
- Problem not differentiated
- "No time to exercise" is the problem of EVERY fitness app
- Existing alternatives: Peloton, Nike Training, YouTube, Apple Fitness+
- Why would someone SWITCH from these?
- Solution is a feature list, not a value prop
- "10-minute workouts" exists everywhere
- "Kid-friendly" - what does this actually mean?
- Revenue assumptions untested
- $9.99/month - based on what?
- Fitness apps have notoriously high churn
- Most people don't pay for fitness apps
Riskiest Assumptions (in order):
| # | Assumption | Risk | Evidence Needed |
|---|------------|------|-----------------|
| 1 | Parents will PAY for yet another fitness app | FATAL | Only 2-5% of app downloaders pay. Need proof of WTP |
| 2 | "Kid-friendly" is a differentiator people care about | HIGH | Does this problem exist? Do they seek solutions? |
| 3 | 10-minute workouts are effective enough | MEDIUM | Will users see results and stay? |
| 4 | $9.99 is the right price | MEDIUM | Could be too high or too low |
Recommendation:
Before building anything, validate:
- Do parents with kids SPECIFICALLY want kid-friendly fitness?
- Interview 15 parents who've TRIED fitness apps
- Ask: What made you quit? What would make you stay?
- Is "kid-friendly" a real differentiator?
- What does this mean? Kids do it with them? Kids don't interrupt?
- Test: Would they pay MORE for this feature?
- Willingness to pay?
- How many of your target users currently pay for any fitness solution?
- If <30%, this market may not monetize
Revised Canvas Hypothesis:
## More Specific Lean Canvas
Customer Segment:
- Primary: Stay-at-home moms, ages 28-40, with kids 2-6
- Early Adopter: Former gym-goers who quit after having kids
and have tried 2+ fitness apps in past year
Problem (validated):
- Can't workout because toddler interrupts/needs attention
- Guilt about "ignoring" kids to exercise
- Previous apps failed because required uninterrupted time
Unique Value Proposition:
"The only workout you can do WITH your toddler, not despite them."
Solution:
- Workouts designed for parent + kid together
- Toddler becomes part of the exercise (lifting, playing)
- 15-min max, no equipment, living room safe
This is more specific and testable.
Example 3: Pivot Comparison
Input:
> "I'm considering pivoting. Compare these two Lean Canvases and recommend which to pursue."
Canvas A: B2B Analytics Dashboard
- Customer: Marketing managers at e-commerce companies
- Problem: Too many analytics tools, no unified view
- Solution: All-in-one dashboard connecting GA, FB, Shopify
- Revenue: $299/month
- Unfair Advantage: None yet
Canvas B: Shopify App for Email Segmentation
- Customer: Shopify store owners doing 100K-1M revenue
- Problem: Mailchimp/Klaviyo segments are basic
- Solution: AI-powered customer segmentation for Shopify
- Revenue: $49/month
- Unfair Advantage: Shopify partnership potential
Output:
Comparative Analysis:
| Dimension | Canvas A (Dashboard) | Canvas B (Shopify App) |
|-----------|---------------------|----------------------|
| Market Size | Large (all e-commerce) | Smaller (Shopify only) |
| Competition | HIGH (Databox, Klipfolio, Geckoboard, custom) | MEDIUM (some apps exist) |
| Differentiation | LOW (everyone does "unified") | MEDIUM (AI segmentation specific) |
| Distribution | Hard (enterprise sales) | EASIER (Shopify App Store) |
| Price Point | $299 (requires sales) | $49 (self-serve possible) |
| Unfair Advantage | None | Shopify ecosystem |
| Time to Revenue | Longer | Shorter |
Risk Assessment:
Canvas A Risks:
| Risk | Level | Notes |
|------|-------|-------|
| Building a commodity | HIGH | "Unified dashboard" = red ocean |
| Long sales cycles | HIGH | $299 needs demos, trials |
| Integration complexity | HIGH | Must support many platforms |
| No defensibility | HIGH | Easy to copy |
Canvas B Risks:
| Risk | Level | Notes |
|------|-------|-------|
| Platform dependency | HIGH | Shopify could change rules |
| Smaller market | MEDIUM | Limited to Shopify stores |
| AI hype skepticism | MEDIUM | Must prove AI actually works better |
| Low price = high volume needed | MEDIUM | Need ~200 customers to be meaningful |
Recommendation: PURSUE CANVAS B
Reasoning:
- Faster validation: Shopify App Store = immediate distribution. You can test in weeks.
- Clearer differentiation: "AI segmentation" is specific vs. "unified dashboard" which is generic.
- Better path to defensibility:
- App Store reviews and ratings compound
- Shopify partner program potential
- Data from customers creates better AI
- Lower risk profile: $49 self-serve is easier to sell than $299 with demos.
- Pivot optionality: If it works on Shopify, you can expand to other platforms. If Canvas A fails, you have nothing.
BUT validate first:
- Is "segmentation" really the pain point? Or is it "I don't know what emails to send"?
- Do Shopify merchants actually buy apps? What's average spend?
- What's the competitive landscape in Shopify App Store?
Checklists & Templates
Lean Canvas Template (Blank)
## LEAN CANVAS: [Product Name]
**Version:** 1.0
**Date:** ___________
**Author:** ___________
┌─────────────────────────────────────────────────────────────────────┐
│ │
│ 2. PROBLEM │ 4. SOLUTION │
│ ──────────── │ ──────────── │
│ 1. │ 1. │
│ 2. │ 2. │
│ 3. │ 3. │
│ │ │
│ Existing Alternatives: ├─────────────────────────────────│
│ - │ 8. KEY METRICS │
│ - │ ──────────────── │
│ │ A: │
│ │ A: │
│ │ R: │
│ │ R: │
│ │ R: │
│ │ OMTM: │
├──────────────────────────────────┴──────────────────────────────────┤
│ 3. UNIQUE VALUE PROPOSITION │
│ ──────────────────────────── │
│ │
│ High-level concept: │
├─────────────────────────────────────────────────────────────────────┤
│ 9. UNFAIR ADVANTAGE │ 5. CHANNELS │ 1. │
│ ──────────────────── │ ──────────── │ CUST │
│ │ - │ SEGS │
│ │ - │ ──── │
│ │ - │ │
│ │ │ │
│ │ │ EA: │
│ │ │ │
├─────────────────────────────────┴───────────────────────────┴──────┤
│ 7. COST STRUCTURE │ 6. REVENUE STREAMS │
│ ────────────────── │ ──────────────────── │
│ Fixed: │ Model: │
│ │ Price: │
│ Variable: │ │
│ │ LTV: │
│ CAC: │ LTV:CAC: │
│ Break-even: │ │
└─────────────────────────────────────┴──────────────────────────────┘
Lean Canvas Review Checklist
## Lean Canvas Quality Check
### Completeness
- [ ] All 9 boxes filled
- [ ] Customer segment is specific (not generic)
- [ ] Early adopters identified
- [ ] Problems are customer problems (not your assumptions)
- [ ] Solution maps to problems
- [ ] UVP is clear in one sentence
- [ ] Metrics are measurable
### Quality
- [ ] Problems validated (or marked as hypothesis)
- [ ] Existing alternatives researched (not guessed)
- [ ] Revenue model makes mathematical sense
- [ ] Costs are realistic
- [ ] Unfair advantage is real (or honestly "none yet")
### Risks Identified
- [ ] Top 3 riskiest assumptions documented
- [ ] Validation experiments designed
- [ ] Go/no-go criteria defined
Lean Canvas Versioning Template
## Lean Canvas Evolution Log
### Version 1.0 - [Date]
Initial hypothesis
Key assumptions: [list]
### Version 1.1 - [Date]
**What changed:** [box(es) updated]
**Why:** [evidence/learning that caused change]
**Key assumptions now:** [updated list]
### Version 2.0 - [Date] (Major Pivot)
**What changed:** [customer/problem/solution pivot]
**Why:** [what invalidated previous version]
**New hypothesis:** [summary]
Skill Boundaries
What This Skill Does Well
- Structuring audio production workflows
- Providing technical guidance
- Creating quality checklists
- Suggesting creative approaches
What This Skill Cannot Do
- Replace audio engineering expertise
- Make subjective creative decisions
- Access or edit audio files directly
- Guarantee commercial success
References
- Maurya, Ash. "Running Lean" (2012) - Original Lean Canvas methodology
- Maurya, Ash. "Scaling Lean" (2016) - Traction roadmap
- Osterwalder, Alex. "Business Model Generation" (2010) - Original BMC
- Blank, Steve. "The Startup Owner's Manual" (2012) - Customer Development
- Ries, Eric. "The Lean Startup" (2011) - Build-Measure-Learn context
Related Skills
- customer-discovery - Methodology to validate canvas boxes
- mom-test - Interview techniques for validation
- jobs-to-be-done - Problem understanding framework
- value-proposition-canvas - Deep dive on customer-solution fit
- first-principles - Challenge assumptions in your canvas
Skill Metadata (Internal Use)
name: lean-canvas
category: validation
subcategory: business-model
version: 1.0
author: MKTG Skills
source_expert: Ash Maurya
source_work: Running Lean
difficulty: beginner
estimated_value: $2,000 startup strategy session
tags: [business-model, validation, startups, YC, lean-startup, canvas]
created: 2026-01-25
updated: 2026-01-25
Other skills for the same job
different authors, same section of the catalogue
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Improves the quality of images, especially screenshots, by enhancing resolution, sharpness, and clarity. Perfect for preparing images for presentations, documentation, or social media posts.
Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.
Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.
Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows.
Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications.
This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.
How to use it
Take guia-matthieu/lean-canvas from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.