Produces the requested working draft
The source is designed to produce the specific draft or structured output described in its instruction, using the information supplied in the conversation.
AdvancedAlireza Rezvani
Experiment Designer
Experiment Designer is an ecommerce AI skill for Alireza Rezvani, built for teams working with Codex, Claude Code, OpenClaw. Use it to improve a product page,…
- Views
- 0
- Installs
- 0
- Favorites
- 0
What this Skill does
You are preparing to improve a product page, shopping journey, or checkout experience. Do not change every step at once: test Planning A/B tests with proper sample size estimation and statistical power alongside Writing testable hypotheses with clear success and failure… Start with a small test around “Planning A/B tests with proper sample size estimation and statistical power”, then check whether “Writing testable hypotheses with clear success and failure criteria” fits the way your team actually works.
What makes it different
Structures the analysis before making a recommendation
The source asks for analysis, classification, ranking, or scoring before it reaches a conclusion or next action.
Before you use it
Python runtime
The source includes a Python command or script. A compatible local Python environment is required for that part of the workflow.
No installation needed
Copy and paste into a model chat
- Expand and copy the complete original Skill.md below.
- Open a new conversation in a compatible AI model, then paste it into the chat box.
- Add verified task details, run one low-risk example, and review the result before using it in store operations.
Original Skill.md
The complete source is shown below. Copy it from the top right to use it.
Design, prioritize, and evaluate product experiments. Follow this workflow: 1) Write hypothesis in If/Then/Because format — If [intervention], Then [metric] changes by [direction/magnitude], Because [mechanism]. 2) Define primary, guardrail, and secondary metrics before testing. 3) Estimate sample size via scripts/sample_size_calculator.py using baseline rate, MDE, alpha, and power. 4) Prioritize with ICE score (Impact × Confidence × Ease / 10). 5) Launch with pre-defined stopping rules; avoid repeated peeking. 6) Interpret results using confidence intervals and practical significance, not just p-values. Guard against underpowered tests, mid-test changes, sample ratio mismatch, and instrumentation drift.Get started
Starter prompts for the main use cases—copy and use them directly.
01Start with one real task
Do not begin with a store-wide rollout. Pick one reversible task where Experiment Designer can help you plan campaigns, produce channel-ready material, and control advertising work. Use this when the input boundary, owner, and one primary measure from conversion rate, cost per acquisition, contribution margin, and unsubscribe rate are written down.
Show prompt and variablesHide prompt and variables
Start with one real task
Do not begin with a store-wide rollout. Pick one reversible task where Experiment Designer can help you plan campaigns, produce channel-ready material, and control advertising work. Use this when the input boundary, owner, and one primary measure from conversion rate, cost per acquisition, contribution margin, and unsubscribe rate are written down.
Use the Skill above to help me with this task: Start with one real task.
Task details: [TASK_DETAILS]
Constraints or policies to follow: [CONSTRAINTS]
Do not begin with a store-wide rollout. Pick one reversible task where Experiment Designer can help you plan campaigns, produce channel-ready material, and control advertising work.
Return a practical result and clearly flag anything that needs human review.Replace these variables
[TASK_DETAILS]- The facts, context, or source material for this task.
[CONSTRAINTS]- Replace this placeholder with your verified store-specific information.
02Prepare the input and guardrails
Collect only the the offer, audience, approved claims, brand voice, channel limits, and budget needed for this test. Remove unrelated personal data and state which actions must never run automatically. Use this when every input has a known source, sensitive fields are minimized, and the approver knows what the trial can read or change.
Show prompt and variablesHide prompt and variables
Prepare the input and guardrails
Collect only the the offer, audience, approved claims, brand voice, channel limits, and budget needed for this test. Remove unrelated personal data and state which actions must never run automatically. Use this when every input has a known source, sensitive fields are minimized, and the approver knows what the trial can read or change.
Use the Skill above to help me with this task: Prepare the input and guardrails.
Task details: [TASK_DETAILS]
Constraints or policies to follow: [CONSTRAINTS]
Collect only the the offer, audience, approved claims, brand voice, channel limits, and budget needed for this test. Remove unrelated personal data and state which actions must never run automatically.
Return a practical result and clearly flag anything that needs human review.Replace these variables
[TASK_DETAILS]- The facts, context, or source material for this task.
[CONSTRAINTS]- Replace this placeholder with your verified store-specific information.
03Inspect the source Skill, then run it
Read the source, installation method, and permission notes before adding Experiment Designer to a separate test project. Keep commands and Skill text exactly as published. Use this when you have a campaign brief or marketing asset that a responsible operator can inspect, and it stayed inside the approved boundary.
Show prompt and variablesHide prompt and variables
Inspect the source Skill, then run it
Read the source, installation method, and permission notes before adding Experiment Designer to a separate test project. Keep commands and Skill text exactly as published. Use this when you have a campaign brief or marketing asset that a responsible operator can inspect, and it stayed inside the approved boundary.
Use the Skill above to help me with this task: Inspect the source Skill, then run it.
Task details: [TASK_DETAILS]
Constraints or policies to follow: [CONSTRAINTS]
Read the source, installation method, and permission notes before adding Experiment Designer to a separate test project. Keep commands and Skill text exactly as published.
Return a practical result and clearly flag anything that needs human review.Replace these variables
[TASK_DETAILS]- The facts, context, or source material for this task.
[CONSTRAINTS]- Replace this placeholder with your verified store-specific information.
04Review it against a baseline
Do not judge the result by fluency. Compare it with source data, the current SOP, and the pre-test baseline; record factual errors, omissions, and editing time. Use this when conversion rate, cost per acquisition, contribution margin, and unsubscribe rate has a pre-test baseline, and errors and exceptions are logged separately.
Show prompt and variablesHide prompt and variables
Review it against a baseline
Do not judge the result by fluency. Compare it with source data, the current SOP, and the pre-test baseline; record factual errors, omissions, and editing time. Use this when conversion rate, cost per acquisition, contribution margin, and unsubscribe rate has a pre-test baseline, and errors and exceptions are logged separately.
Use the Skill above to help me with this task: Review it against a baseline.
Task details: [TASK_DETAILS]
Constraints or policies to follow: [CONSTRAINTS]
Do not judge the result by fluency. Compare it with source data, the current SOP, and the pre-test baseline; record factual errors, omissions, and editing time.
Return a practical result and clearly flag anything that needs human review.Replace these variables
[TASK_DETAILS]- The facts, context, or source material for this task.
[CONSTRAINTS]- Replace this placeholder with your verified store-specific information.
Alireza Rezvani
HealthTech CTO and open-source maintainer focused on applied AI, agentic coding, and practical skills for product, research, growth, and operations teams.
Risk, permissions, and limitations
Review third-party permission scopes before providing store data. Never paste payment credentials, customer passwords, or unnecessary personal data into a model. Outputs must be checked by the operator responsible for the workflow.
Content checked:
No reviews yet.