[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6ADZYSDZ5VBXNEF99239G3A":3,"skill-same:01M6ADZYSDZ5VBXNEF99239G3A":103},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M6ADZYSDZ5VBXNEF99239G3A","ab-test-setup",null,"When the user wants to plan, design, or implement an A\u002FB test or experiment. Also use when the user mentions \"A\u002FB test,\" \"split test,\" \"experiment,\" \"test this change,\" \"variant copy,\" \"multivariate test,\" or \"hypothesis.\" For tracking implementation, see analytics-tracking.","davila7\u002Fclaude-code-templates","davila7","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6216945?v=4","01M6ADEWWFPS191NFM0Y1EX1JH","Daniel Avila","01M6ADJHGM78PHQRD3XZTFDVKK",33481,"MIT","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fbusiness-marketing\u002Fab-test-setup\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"test","ab test","setup","ab test setup","ab","test setup","experiment","when","variant","to plan","for","split","implementation","design","use","plan","hypothesis","analytics","change","implement","2026-11-10 10:12:30",false,"---\nname: ab-test-setup\ndescription: When the user wants to plan, design, or implement an A\u002FB test or experiment. Also use when the user mentions \"A\u002FB test,\" \"split test,\" \"experiment,\" \"test this change,\" \"variant copy,\" \"multivariate test,\" or \"hypothesis.\" For tracking implementation, see analytics-tracking.\n---\n\n# A\u002FB Test Setup\n\nYou are an expert in experimentation and A\u002FB testing. Your goal is to help design tests that produce statistically valid, actionable results.\n\n## Initial Assessment\n\nBefore designing a test, understand:\n\n1. **Test Context**\n   - What are you trying to improve?\n   - What change are you considering?\n   - What made you want to test this?\n\n2. **Current State**\n   - Baseline conversion rate?\n   - Current traffic volume?\n   - Any historical test data?\n\n3. **Constraints**\n   - Technical implementation complexity?\n   - Timeline requirements?\n   - Tools available?\n\n---\n\n## Core Principles\n\n### 1. Start with a Hypothesis\n- Not just \"let's see what happens\"\n- Specific prediction of outcome\n- Based on reasoning or data\n\n### 2. Test One Thing\n- Single variable per test\n- Otherwise you don't know what worked\n- Save MVT for later\n\n### 3. Statistical Rigor\n- Pre-determine sample size\n- Don't peek and stop early\n- Commit to the methodology\n\n### 4. Measure What Matters\n- Primary metric tied to business value\n- Secondary metrics for context\n- Guardrail metrics to prevent harm\n\n---\n\n## Hypothesis Framework\n\n### Structure\n\n```\nBecause [observation\u002Fdata],\nwe believe [change]\nwill cause [expected outcome]\nfor [audience].\nWe'll know this is true when [metrics].\n```\n\n### Examples\n\n**Weak hypothesis:**\n\"Changing the button color might increase clicks.\"\n\n**Strong hypothesis:**\n\"Because users report difficulty finding the CTA (per heatmaps and feedback), we believe making the button larger and using contrasting color will increase CTA clicks by 15%+ for new visitors. We'll measure click-through rate from page view to signup start.\"\n\n### Good Hypotheses Include\n\n- **Observation**: What prompted this idea\n- **Change**: Specific modification\n- **Effect**: Expected outcome and direction\n- **Audience**: Who this applies to\n- **Metric**: How you'll measure success\n\n---\n\n## Test Types\n\n### A\u002FB Test (Split Test)\n- Two versions: Control (A) vs. Variant (B)\n- Single change between versions\n- Most common, easiest to analyze\n\n### A\u002FB\u002Fn Test\n- Multiple variants (A vs. B vs. C...)\n- Requires more traffic\n- Good for testing several options\n\n### Multivariate Test (MVT)\n- Multiple changes in combinations\n- Tests interactions between changes\n- Requires significantly more traffic\n- Complex analysis\n\n### Split URL Test\n- Different URLs for variants\n- Good for major page changes\n- Easier implementation sometimes\n\n---\n\n## Sample Size Calculation\n\n### Inputs Needed\n\n1. **Baseline conversion rate**: Your current rate\n2. **Minimum detectable effect (MDE)**: Smallest change worth detecting\n3. **Statistical significance level**: Usually 95%\n4. **Statistical power**: Usually 80%\n\n### Quick Reference\n\n| Baseline Rate | 10% Lift | 20% Lift | 50% Lift |\n|---------------|----------|----------|----------|\n| 1% | 150k\u002Fvariant | 39k\u002Fvariant | 6k\u002Fvariant |\n| 3% | 47k\u002Fvariant | 12k\u002Fvariant | 2k\u002Fvariant |\n| 5% | 27k\u002Fvariant | 7k\u002Fvariant | 1.2k\u002Fvariant |\n| 10% | 12k\u002Fvariant | 3k\u002Fvariant | 550\u002Fvariant |\n\n### Formula Resources\n- Evan Miller's calculator: https:\u002F\u002Fwww.evanmiller.org\u002Fab-testing\u002Fsample-size.html\n- Optimizely's calculator: https:\u002F\u002Fwww.optimizely.com\u002Fsample-size-calculator\u002F\n\n### Test Duration\n\n```\nDuration = Sample size needed per variant × Number of variants\n           ───────────────────────────────────────────────────\n           Daily traffic to test page × Conversion rate\n```\n\nMinimum: 1-2 business cycles (usually 1-2 weeks)\nMaximum: Avoid running too long (novelty effects, external factors)\n\n---\n\n## Metrics Selection\n\n### Primary Metric\n- Single metric that matters most\n- Directly tied to hypothesis\n- What you'll use to call the test\n\n### Secondary Metrics\n- Support primary metric interpretation\n- Explain why\u002Fhow the change worked\n- Help understand user behavior\n\n### Guardrail Metrics\n- Things that shouldn't get worse\n- Revenue, retention, satisfaction\n- Stop test if significantly negative\n\n### Metric Examples by Test Type\n\n**Homepage CTA test:**\n- Primary: CTA click-through rate\n- Secondary: Time to click, scroll depth\n- Guardrail: Bounce rate, downstream conversion\n\n**Pricing page test:**\n- Primary: Plan selection rate\n- Secondary: Time on page, plan distribution\n- Guardrail: Support tickets, refund rate\n\n**Signup flow test:**\n- Primary: Signup completion rate\n- Secondary: Field-level completion, time to complete\n- Guardrail: User activation rate (post-signup quality)\n\n---\n\n## Designing Variants\n\n### Control (A)\n- Current experience, unchanged\n- Don't modify during test\n\n### Variant (B+)\n\n**Best practices:**\n- Single, meaningful change\n- Bold enough to make a difference\n- True to the hypothesis\n\n**What to vary:**\n\nHeadlines\u002FCopy:\n- Message angle\n- Value proposition\n- Specificity level\n- Tone\u002Fvoice\n\nVisual Design:\n- Layout structure\n- Color and contrast\n- Image selection\n- Visual hierarchy\n\nCTA:\n- Button copy\n- Size\u002Fprominence\n- Placement\n- Number of CTAs\n\nContent:\n- Information included\n- Order of information\n- Amount of content\n- Social proof type\n\n### Documenting Variants\n\n```\nControl (A):\n- Screenshot\n- Description of current state\n\nVariant (B):\n- Screenshot or mockup\n- Specific changes made\n- Hypothesis for why this will win\n```\n\n---\n\n## Traffic Allocation\n\n### Standard Split\n- 50\u002F50 for A\u002FB test\n- Equal split for multiple variants\n\n### Conservative Rollout\n- 90\u002F10 or 80\u002F20 initially\n- Limits risk of bad variant\n- Longer to reach significance\n\n### Ramping\n- Start small, increase over time\n- Good for technical risk mitigation\n- Most tools support this\n\n### Considerations\n- Consistency: Users see same variant on return\n- Segment sizes: Ensure segments are large enough\n- Time of day\u002Fweek: Balanced exposure\n\n---\n\n## Implementation Approaches\n\n### Client-Side Testing\n\n**Tools**: PostHog, Optimizely, VWO, custom\n\n**How it works**:\n- JavaScript modifies page after load\n- Quick to implement\n- Can cause flicker\n\n**Best for**:\n- Marketing pages\n- Copy\u002Fvisual changes\n- Quick iteration\n\n### Server-Side Testing\n\n**Tools**: PostHog, LaunchDarkly, Split, custom\n\n**How it works**:\n- Variant determined before page renders\n- No flicker\n- Requires development work\n\n**Best for**:\n- Product features\n- Complex changes\n- Performance-sensitive pages\n\n### Feature Flags\n\n- Binary on\u002Foff (not true A\u002FB)\n- Good for rollouts\n- Can convert to A\u002FB with percentage split\n\n---\n\n## Running the Test\n\n### Pre-Launch Checklist\n\n- [ ] Hypothesis documented\n- [ ] Primary metric defined\n- [ ] Sample size calculated\n- [ ] Test duration estimated\n- [ ] Variants implemented correctly\n- [ ] Tracking verified\n- [ ] QA completed on all variants\n- [ ] Stakeholders informed\n\n### During the Test\n\n**DO:**\n- Monitor for technical issues\n- Check segment quality\n- Document any external factors\n\n**DON'T:**\n- Peek at results and stop early\n- Make changes to variants\n- Add traffic from new sources\n- End early because you \"know\" the answer\n\n### Peeking Problem\n\nLooking at results before reaching sample size and stopping when you see significance leads to:\n- False positives\n- Inflated effect sizes\n- Wrong decisions\n\n**Solutions:**\n- Pre-commit to sample size and stick to it\n- Use sequential testing if you must peek\n- Trust the process\n\n---\n\n## Analyzing Results\n\n### Statistical Significance\n\n- 95% confidence = p-value \u003C 0.05\n- Means: \u003C5% chance result is random\n- Not a guarantee—just a threshold\n\n### Practical Significance\n\nStatistical ≠ Practical\n\n- Is the effect size meaningful for business?\n- Is it worth the implementation cost?\n- Is it sustainable over time?\n\n### What to Look At\n\n1. **Did you reach sample size?**\n   - If not, result is preliminary\n\n2. **Is it statistically significant?**\n   - Check confidence intervals\n   - Check p-value\n\n3. **Is the effect size meaningful?**\n   - Compare to your MDE\n   - Project business impact\n\n4. **Are secondary metrics consistent?**\n   - Do they support the primary?\n   - Any unexpected effects?\n\n5. **Any guardrail concerns?**\n   - Did anything get worse?\n   - Long-term risks?\n\n6. **Segment differences?**\n   - Mobile vs. desktop?\n   - New vs. returning?\n   - Traffic source?\n\n### Interpreting Results\n\n| Result | Conclusion |\n|--------|------------|\n| Significant winner | Implement variant |\n| Significant loser | Keep control, learn why |\n| No significant difference | Need more traffic or bolder test |\n| Mixed signals | Dig deeper, maybe segment |\n\n---\n\n## Documenting and Learning\n\n### Test Documentation\n\n```\nTest Name: [Name]\nTest ID: [ID in testing tool]\nDates: [Start] - [End]\nOwner: [Name]\n\nHypothesis:\n[Full hypothesis statement]\n\nVariants:\n- Control: [Description + screenshot]\n- Variant: [Description + screenshot]\n\nResults:\n- Sample size: [achieved vs. target]\n- Primary metric: [control] vs. [variant] ([% change], [confidence])\n- Secondary metrics: [summary]\n- Segment insights: [notable differences]\n\nDecision: [Winner\u002FLoser\u002FInconclusive]\nAction: [What we're doing]\n\nLearnings:\n[What we learned, what to test next]\n```\n\n### Building a Learning Repository\n\n- Central location for all tests\n- Searchable by page, element, outcome\n- Prevents re-running failed tests\n- Builds institutional knowledge\n\n---\n\n## Output Format\n\n### Test Plan Document\n\n```\n# A\u002FB Test: [Name]\n\n## Hypothesis\n[Full hypothesis using framework]\n\n## Test Design\n- Type: A\u002FB \u002F A\u002FB\u002Fn \u002F MVT\n- Duration: X weeks\n- Sample size: X per variant\n- Traffic allocation: 50\u002F50\n\n## Variants\n[Control and variant descriptions with visuals]\n\n## Metrics\n- Primary: [metric and definition]\n- Secondary: [list]\n- Guardrails: [list]\n\n## Implementation\n- Method: Client-side \u002F Server-side\n- Tool: [Tool name]\n- Dev requirements: [If any]\n\n## Analysis Plan\n- Success criteria: [What constitutes a win]\n- Segment analysis: [Planned segments]\n```\n\n### Results Summary\nWhen test is complete\n\n### Recommendations\nNext steps based on results\n\n---\n\n## Common Mistakes\n\n### Test Design\n- Testing too small a change (undetectable)\n- Testing too many things (can't isolate)\n- No clear hypothesis\n- Wrong audience\n\n### Execution\n- Stopping early\n- Changing things mid-test\n- Not checking implementation\n- Uneven traffic allocation\n\n### Analysis\n- Ignoring confidence intervals\n- Cherry-picking segments\n- Over-interpreting inconclusive results\n- Not considering practical significance\n\n---\n\n## Questions to Ask\n\nIf you need more context:\n1. What's your current conversion rate?\n2. How much traffic does this page get?\n3. What change are you considering and why?\n4. What's the smallest improvement worth detecting?\n5. What tools do you have for testing?\n6. Have you tested this area before?\n\n---\n\n## Related Skills\n\n- **page-cro**: For generating test ideas based on CRO principles\n- **analytics-tracking**: For setting up test measurement\n- **copywriting**: For creating variant copy\n","# A\u002FB Test Setup\n\nYou are an expert in experimentation and A\u002FB testing. Your goal is to help design tests that produce statistically valid, actionable results.\n\n## Initial Assessment\n\nBefore designing a test, understand:\n\n1. **Test Context**\n   - What are you trying to improve?\n   - What change are you considering?\n   - What made you want to test this?\n\n2. **Current State**\n   - Baseline conversion rate?\n   - Current traffic volume?\n   - Any historical test data?\n\n3. **Constraints**\n   - Technical implementation complexity?\n   - Timeline requirements?\n   - Tools available?\n\n---\n\n## Core Principles\n\n### 1. Start with a Hypothesis\n- Not just \"let's see what happens\"\n- Specific prediction of outcome\n- Based on reasoning or data\n\n### 2. Test One Thing\n- Single variable per test\n- Otherwise you don't know what worked\n- Save MVT for later\n\n### 3. Statistical Rigor\n- Pre-determine sample size\n- Don't peek and stop early\n- Commit to the methodology\n\n### 4. Measure What Matters\n- Primary metric tied to business value\n- Secondary metrics for context\n- Guardrail metrics to prevent harm\n\n---\n\n## Hypothesis Framework\n\n### Structure\n\n```\nBecause [observation\u002Fdata],\nwe believe [change]\nwill cause [expected outcome]\nfor [audience].\nWe'll know this is true when [metrics].\n```\n\n### Examples\n\n**Weak hypothesis:**\n\"Changing the button color might increase clicks.\"\n\n**Strong hypothesis:**\n\"Because users report difficulty finding the CTA (per heatmaps and feedback), we believe making the button larger and using contrasting color will increase CTA clicks by 15%+ for new visitors. We'll measure click-through rate from page view to signup start.\"\n\n### Good Hypotheses Include\n\n- **Observation**: What prompted this idea\n- **Change**: Specific modification\n- **Effect**: Expected outcome and direction\n- **Audience**: Who this applies to\n- **Metric**: How you'll measure success\n\n---\n\n## Test Types\n\n### A\u002FB Test (Split Test)\n- Two versions: Control (A) vs. Variant (B)\n- Single change between versions\n- Most common, easiest to analyze\n\n### A\u002FB\u002Fn Test\n- Multiple variants (A vs. B vs. C...)\n- Requires more traffic\n- Good for testing several options\n\n### Multivariate Test (MVT)\n- Multiple changes in combinations\n- Tests interactions between changes\n- Requires significantly more traffic\n- Complex analysis\n\n### Split URL Test\n- Different URLs for variants\n- Good for major page changes\n- Easier implementation sometimes\n\n---\n\n## Sample Size Calculation\n\n### Inputs Needed\n\n1. **Baseline conversion rate**: Your current rate\n2. **Minimum detectable effect (MDE)**: Smallest change worth detecting\n3. **Statistical significance level**: Usually 95%\n4. **Statistical power**: Usually 80%\n\n### Quick Reference\n\n| Baseline Rate | 10% Lift | 20% Lift | 50% Lift |\n|---------------|----------|----------|----------|\n| 1% | 150k\u002Fvariant | 39k\u002Fvariant | 6k\u002Fvariant |\n| 3% | 47k\u002Fvariant | 12k\u002Fvariant | 2k\u002Fvariant |\n| 5% | 27k\u002Fvariant | 7k\u002Fvariant | 1.2k\u002Fvariant |\n| 10% | 12k\u002Fvariant | 3k\u002Fvariant | 550\u002Fvariant |\n\n### Formula Resources\n- Evan Miller's calculator: https:\u002F\u002Fwww.evanmiller.org\u002Fab-testing\u002Fsample-size.html\n- Optimizely's calculator: https:\u002F\u002Fwww.optimizely.com\u002Fsample-size-calculator\u002F\n\n### Test Duration\n\n```\nDuration = Sample size needed per variant × Number of variants\n           ───────────────────────────────────────────────────\n           Daily traffic to test page × Conversion rate\n```\n\nMinimum: 1-2 business cycles (usually 1-2 weeks)\nMaximum: Avoid running too long (novelty effects, external factors)\n\n---\n\n## Metrics Selection\n\n### Primary Metric\n- Single metric that matters most\n- Directly tied to hypothesis\n- What you'll use to call the test\n\n### Secondary Metrics\n- Support primary metric interpretation\n- Explain why\u002Fhow the change worked\n- Help understand user behavior\n\n### Guardrail Metrics\n- Things that shouldn't get worse\n- Revenue, retention, satisfaction\n- Stop test if significantly negative\n\n### Metric Examples by Test Type\n\n**Homepage CTA test:**\n- Primary: CTA click-through rate\n- Secondary: Time to click, scroll depth\n- Guardrail: Bounce rate, downstream conversion\n\n**Pricing page test:**\n- Primary: Plan selection rate\n- Secondary: Time on page, plan distribution\n- Guardrail: Support tickets, refund rate\n\n**Signup flow test:**\n- Primary: Signup completion rate\n- Secondary: Field-level completion, time to complete\n- Guardrail: User activation rate (post-signup quality)\n\n---\n\n## Designing Variants\n\n### Control (A)\n- Current experience, unchanged\n- Don't modify during test\n\n### Variant (B+)\n\n**Best practices:**\n- Single, meaningful change\n- Bold enough to make a difference\n- True to the hypothesis\n\n**What to vary:**\n\nHeadlines\u002FCopy:\n- Message angle\n- Value proposition\n- Specificity level\n- Tone\u002Fvoice\n\nVisual Design:\n- Layout structure\n- Color and contrast\n- Image selection\n- Visual hierarchy\n\nCTA:\n- Button copy\n- Size\u002Fprominence\n- Placement\n- Number of CTAs\n\nContent:\n- Information included\n- Order of information\n- Amount of content\n- Social proof type\n\n### Documenting Variants\n\n```\nControl (A):\n- Screenshot\n- Description of current state\n\nVariant (B):\n- Screenshot or mockup\n- Specific changes made\n- Hypothesis for why this will win\n```\n\n---\n\n## Traffic Allocation\n\n### Standard Split\n- 50\u002F50 for A\u002FB test\n- Equal split for multiple variants\n\n### Conservative Rollout\n- 90\u002F10 or 80\u002F20 initially\n- Limits risk of bad variant\n- Longer to reach significance\n\n### Ramping\n- Start small, increase over time\n- Good for technical risk mitigation\n- Most tools support this\n\n### Considerations\n- Consistency: Users see same variant on return\n- Segment sizes: Ensure segments are large enough\n- Time of day\u002Fweek: Balanced exposure\n\n---\n\n## Implementation Approaches\n\n### Client-Side Testing\n\n**Tools**: PostHog, Optimizely, VWO, custom\n\n**How it works**:\n- JavaScript modifies page after load\n- Quick to implement\n- Can cause flicker\n\n**Best for**:\n- Marketing pages\n- Copy\u002Fvisual changes\n- Quick iteration\n\n### Server-Side Testing\n\n**Tools**: PostHog, LaunchDarkly, Split, custom\n\n**How it works**:\n- Variant determined before page renders\n- No flicker\n- Requires development work\n\n**Best for**:\n- Product features\n- Complex changes\n- Performance-sensitive pages\n\n### Feature Flags\n\n- Binary on\u002Foff (not true A\u002FB)\n- Good for rollouts\n- Can convert to A\u002FB with percentage split\n\n---\n\n## Running the Test\n\n### Pre-Launch Checklist\n\n- [ ] Hypothesis documented\n- [ ] Primary metric defined\n- [ ] Sample size calculated\n- [ ] Test duration estimated\n- [ ] Variants implemented correctly\n- [ ] Tracking verified\n- [ ] QA completed on all variants\n- [ ] Stakeholders informed\n\n### During the Test\n\n**DO:**\n- Monitor for technical issues\n- Check segment quality\n- Document any external factors\n\n**DON'T:**\n- Peek at results and stop early\n- Make changes to variants\n- Add traffic from new sources\n- End early because you \"know\" the answer\n\n### Peeking Problem\n\nLooking at results before reaching sample size and stopping when you see significance leads to:\n- False positives\n- Inflated effect sizes\n- Wrong decisions\n\n**Solutions:**\n- Pre-commit to sample size and stick to it\n- Use sequential testing if you must peek\n- Trust the process\n\n---\n\n## Analyzing Results\n\n### Statistical Significance\n\n- 95% confidence = p-value \u003C 0.05\n- Means: \u003C5% chance result is random\n- Not a guarantee—just a threshold\n\n### Practical Significance\n\nStatistical ≠ Practical\n\n- Is the effect size meaningful for business?\n- Is it worth the implementation cost?\n- Is it sustainable over time?\n\n### What to Look At\n\n1. **Did you reach sample size?**\n   - If not, result is preliminary\n\n2. **Is it statistically significant?**\n   - Check confidence intervals\n   - Check p-value\n\n3. **Is the effect size meaningful?**\n   - Compare to your MDE\n   - Project business impact\n\n4. **Are secondary metrics consistent?**\n   - Do they support the primary?\n   - Any unexpected effects?\n\n5. **Any guardrail concerns?**\n   - Did anything get worse?\n   - Long-term risks?\n\n6. **Segment differences?**\n   - Mobile vs. desktop?\n   - New vs. returning?\n   - Traffic source?\n\n### Interpreting Results\n\n| Result | Conclusion |\n|--------|------------|\n| Significant winner | Implement variant |\n| Significant loser | Keep control, learn why |\n| No significant difference | Need more traffic or bolder test |\n| Mixed signals | Dig deeper, maybe segment |\n\n---\n\n## Documenting and Learning\n\n### Test Documentation\n\n```\nTest Name: [Name]\nTest ID: [ID in testing tool]\nDates: [Start] - [End]\nOwner: [Name]\n\nHypothesis:\n[Full hypothesis statement]\n\nVariants:\n- Control: [Description + screenshot]\n- Variant: [Description + screenshot]\n\nResults:\n- Sample size: [achieved vs. target]\n- Primary metric: [control] vs. [variant] ([% change], [confidence])\n- Secondary metrics: [summary]\n- Segment insights: [notable differences]\n\nDecision: [Winner\u002FLoser\u002FInconclusive]\nAction: [What we're doing]\n\nLearnings:\n[What we learned, what to test next]\n```\n\n### Building a Learning Repository\n\n- Central location for all tests\n- Searchable by page, element, outcome\n- Prevents re-running failed tests\n- Builds institutional knowledge\n\n---\n\n## Output Format\n\n### Test Plan Document\n\n```\n# A\u002FB Test: [Name]\n\n## Hypothesis\n[Full hypothesis using framework]\n\n## Test Design\n- Type: A\u002FB \u002F A\u002FB\u002Fn \u002F MVT\n- Duration: X weeks\n- Sample size: X per variant\n- Traffic allocation: 50\u002F50\n\n## Variants\n[Control and variant descriptions with visuals]\n\n## Metrics\n- Primary: [metric and definition]\n- Secondary: [list]\n- Guardrails: [list]\n\n## Implementation\n- Method: Client-side \u002F Server-side\n- Tool: [Tool name]\n- Dev requirements: [If any]\n\n## Analysis Plan\n- Success criteria: [What constitutes a win]\n- Segment analysis: [Planned segments]\n```\n\n### Results Summary\nWhen test is complete\n\n### Recommendations\nNext steps based on results\n\n---\n\n## Common Mistakes\n\n### Test Design\n- Testing too small a change (undetectable)\n- Testing too many things (can't isolate)\n- No clear hypothesis\n- Wrong audience\n\n### Execution\n- Stopping early\n- Changing things mid-test\n- Not checking implementation\n- Uneven traffic allocation\n\n### Analysis\n- Ignoring confidence intervals\n- Cherry-picking segments\n- Over-interpreting inconclusive results\n- Not considering practical significance\n\n---\n\n## Questions to Ask\n\nIf you need more context:\n1. What's your current conversion rate?\n2. How much traffic does this page get?\n3. What change are you considering and why?\n4. What's the smallest improvement worth detecting?\n5. What tools do you have for testing?\n6. Have you tested this area before?\n\n---\n\n## Related Skills\n\n- **page-cro**: For generating test ideas based on CRO principles\n- **analytics-tracking**: For setting up test measurement\n- **copywriting**: For creating variant copy\n",{"name":5,"description":7},"main","2b749c82b5d11fac39d9a6e021ce3f5b8a541bd1","cli-tool\u002Fcomponents\u002Fskills\u002Fbusiness-marketing\u002Fab-test-setup",3931,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates","https:\u002F\u002Faitmpl.com","https:\u002F\u002Fgithub.com\u002Fdavila7","https:\u002F\u002Fdanielavila.me","Building AI dev tools with LLMs","2026-11-14 06:14:46",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add davila7\u002Fclaude-code-templates --skill ab-test-setup",{"tool":65,"command":66},"bunx","bunx skills add davila7\u002Fclaude-code-templates --skill ab-test-setup",{"tool":68,"command":69},"pnpm","pnpm dlx skills add davila7\u002Fclaude-code-templates --skill ab-test-setup","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fbusiness-marketing\u002Fab-test-setup\u002FSKILL.md",true,[73,79,83,87,91,95,99],{"path":74,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":77,"sourceHtmlUrl":78},"elements-of-style\u002F01-introductory.md","REFERENCE","参考文档",303,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Felements-of-style\u002F01-introductory.md",{"path":80,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":81,"sourceHtmlUrl":82},"elements-of-style\u002F02-elementary-rules-of-usage.md",12161,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Felements-of-style\u002F02-elementary-rules-of-usage.md",{"path":84,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":85,"sourceHtmlUrl":86},"elements-of-style\u002F03-elementary-principles-of-composition.md",33622,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Felements-of-style\u002F03-elementary-principles-of-composition.md",{"path":88,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":89,"sourceHtmlUrl":90},"elements-of-style\u002F04-a-few-matters-of-form.md",4918,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Felements-of-style\u002F04-a-few-matters-of-form.md",{"path":92,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":93,"sourceHtmlUrl":94},"elements-of-style\u002F05-words-and-expressions-commonly-misused.md",22310,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Felements-of-style\u002F05-words-and-expressions-commonly-misused.md",{"path":96,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":97,"sourceHtmlUrl":98},"README.md",5853,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002FREADME.md",{"path":100,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":101,"sourceHtmlUrl":102},"signs-of-ai-writing.md",94774,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fenterprise-communication\u002Fwriting-clearly-and-concisely\u002Fsigns-of-ai-writing.md",[104,131,157,190,225,247],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":113,"publishedAt":44},"01M70NHBTQ8BV0A2SCAZXEPHCX","wealthville-pool-scoring","Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter\u002FHold\u002FExit\u002FReduce\u002FAvoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fweb-data\u002Fwealthville-pool-scoring\u002FSKILL.md",[110],{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[114,115,116,117,118,34,119,120,38,31,121,122,123,124,125,126,127,128,129,130],"scoring","pool","wealthville pool scoring","liquidity","which","exit","now","reduce","with","record","track","verdict","evm","solana","how","public","look",{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":138,"publishedAt":44},"01M70MB6VR8ERJ4EM8D5KHBMEV","technical-debt-auditor","Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO\u002FFIXME\u002FHACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like \"where should I focus refactoring\", \"find code smells\", \"audit my codebase\", \"what technical debt do we have\", \"prioritize cleanup work\", or asks about code quality metrics. Also trigger when the user mentions \"code rot\", \"legacy cleanup\", \"refactoring priorities\", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say \"technical debt\" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdevelopment\u002Ftechnical-debt-auditor\u002FSKILL.md",[137],{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},[139,140,141,142,143,144,145,146,147,148,149,31,150,151,152,39,153,154,155,156],"technical debt","debt","technical","technical debt auditor","auditor","coverage","complexity","markers","project","cleanup","unused","code","todo","work","focus","signals","audit","dependency",{"skillNo":158,"name":159,"nameZh":6,"description":160,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":161,"categories":162,"keywords":172,"publishedAt":44},"01M70KPYH9QNHCNME5GJQGE22K","readonly-sql-guard","Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production\u002FERP\u002Fanalytics database (a \"query\" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write\u002FDDL\u002Fside-effect\u002Ftransaction-escape risks. Checks what a statement CALLS, not just its shape.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdatabase\u002Freadonly-sql-guard\u002FSKILL.md",[163,166,167],{"categoryNo":164,"categoryName":165,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":168,"categoryName":169,"taxonomyType":170,"taxonomyTypeDesc":171},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[173,174,175,176,177,178,179,34,180,181,182,183,184,185,186,38,187,188,31,189],"readonly","guard","sql guard","sql","readonly sql guard","erp","effect","read-only","db","just","tool","calls","only","MCP","access","database","query",{"skillNo":191,"name":192,"nameZh":6,"description":193,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":194,"licenseFileUrl":195,"sourceHtmlUrl":196,"categories":197,"keywords":201,"publishedAt":44},"01M6AFY6F7CC2BBZP1QD6VTNVC","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FSKILL.md",[198],{"categoryNo":199,"categoryName":200,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223,34,224],"artifacts builder","web","web artifacts builder","artifacts","builder","state","routing","state management","component","css","frontend","ui","Tools","file","components","shadcn","react","Tailwind","suite","jsx","html","single-file","ui components",{"skillNo":226,"name":227,"nameZh":6,"description":228,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":194,"licenseFileUrl":229,"sourceHtmlUrl":230,"categories":231,"keywords":234,"publishedAt":44},"01M6AFWF33SFV9Z3SF4PEEE4BF","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FSKILL.md",[232,233],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":199,"categoryName":200,"taxonomyType":21,"taxonomyTypeDesc":22},[235,236,237,238,239,203,240,212,213,241,242,243,244,34,245,122,246],"webapp testing","testing","webapp","browser","playwright","applications","debugging","toolkit","logs","screenshots","local","behavior",{"skillNo":248,"name":249,"nameZh":6,"description":250,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":194,"licenseFileUrl":251,"sourceHtmlUrl":252,"categories":253,"keywords":261,"publishedAt":44},"01M6AF67Q67JJJ5V51JR2P52KA","discernment-nudge","After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FSKILL.md",[254,257,258],{"categoryNo":255,"categoryName":256,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":259,"categoryName":260,"taxonomyType":170,"taxonomyTypeDesc":171},"01JBSEEDCAT000000000000036","内容创作者",[262,263,264,265,266,267,268,269,205,270,215,150,271,272,273,274,275,185,276,31,277,278,279,280,281,128,282],"discernment nudge","nudge","interpretation","help","run","reasoning","data","writing","lookup","content","review","conversation","short","up","invoke","you","analysis","claims","argument","assemble","recommendations"]