Design and evaluate paid-ad experiments with hypotheses

Paid Media Experiment

by AgriciDaniel · MIT license · ★ 9,466 on GitHub · updated Sep 18

Design and evaluate paid-ad experiments with hypotheses, randomization units, sample-size and duration assumptions, guardrails, platform experiment tools, analysis, and decision rules. Use for A/B test, split test, experiment design,…

---
name: ads-test
description: "Design and evaluate paid-ad experiments with hypotheses, randomization units, sample-size and duration assumptions, guardrails, platform experiment tools, analysis, and decision rules. Use for A/B test, split test, experiment design, hypothesis, statistical significance, sample size, test duration, or experiment readout."
---

# Paid Media Experiment

1. State the decision, causal hypothesis, treatment, control, randomization unit,
   population, primary metric, guardrails, minimum effect, and stopping rule.
2. Check platform constraints, overlapping experiments, conversion lag, seasonality,
   interference, and measurement quality.
3. Calculate sample and duration from declared assumptions; disclose approximations.
4. Change one decision surface unless the design explicitly estimates interactions.
5. Pre-register exclusions, quality checks, analysis, and decision thresholds.
6. For readout, verify assignment integrity and data completeness before estimating
   effect and uncertainty.
7. Return setup or readout in versioned JSON with a plain-language decision.

Overview

Do not repeatedly peek and stop on a favorable result, call underpowered noise a winner, or generalize beyond the tested population.

Manual install (npx, ZIP)

Claude Codenpx skills add AgriciDaniel/claude-ads --skill ads-test -g -a claude-code

Claude Code (manual)

mkdir -p ~/.claude/skills && curl -sL https://codeload.github.com/AgriciDaniel/claude-ads/tar.gz/ac2164493391 \
  | tar -xz -C ~/.claude/skills --strip-components=2 claude-ads-ac2164493391/skills/ads-test

Windows (PowerShell)

New-Item -ItemType Directory -Force "$HOME\.claude\skills" | Out-Null; Invoke-WebRequest "https://m.agentalley.io/zip/e0e60aee3578af3d3df9/ads-test.zip" -OutFile "$env:TEMP\ads-test.zip"; Expand-Archive "$env:TEMP\ads-test.zip" -DestinationPath "$HOME\.claude\skills" -Force

Claude.ai / desktopDownload the ZIP and upload it under Settings → Capabilities → Skills.

Codex, Cursor, other agentsPoint them at SKILL.md. The other files are plain text.

Connects with Agent Alley MCP · agentalley.io/mcp

Files included

SKILL.md1.2 KB

Read the files here

AgriciDaniel/ac216441 file

SKILL.md1.2 KB

SKILL.md

Show the full text22 lines

ads-test/SKILL.md22 lines · 1.2 KB

PreviewCodeRawView on GitHub

Paid Media Experiment

State the decision, causal hypothesis, treatment, control, randomization unit, population, primary metric, guardrails, minimum effect, and stopping rule.
Check platform constraints, overlapping experiments, conversion lag, seasonality, interference, and measurement quality.
Calculate sample and duration from declared assumptions; disclose approximations.
Change one decision surface unless the design explicitly estimates interactions.
Pre-register exclusions, quality checks, analysis, and decision thresholds.
For readout, verify assignment integrity and data completeness before estimating effect and uncertainty.
Return setup or readout in versioned JSON with a plain-language decision.

Do not repeatedly peek and stop on a favorable result, call underpowered noise a winner, or generalize beyond the tested population.

1	---
2	name: ads-test
3	description: "Design and evaluate paid-ad experiments with hypotheses, randomization units, sample-size and duration assumptions, guardrails, platform experiment tools, analysis, and decision rules. Use for A/B test, split test, experiment design, hypothesis, statistical significance, sample size, test duration, or experiment readout."
4	---
5
6	# Paid Media Experiment
7
8	1. State the decision, causal hypothesis, treatment, control, randomization unit,
9	population, primary metric, guardrails, minimum effect, and stopping rule.
10	2. Check platform constraints, overlapping experiments, conversion lag, seasonality,
11	interference, and measurement quality.
12	3. Calculate sample and duration from declared assumptions; disclose approximations.
13	4. Change one decision surface unless the design explicitly estimates interactions.
14	5. Pre-register exclusions, quality checks, analysis, and decision thresholds.
15	6. For readout, verify assignment integrity and data completeness before estimating
16	effect and uncertainty.
17	7. Return setup or readout in versioned JSON with a plain-language decision.
18
19	Do not repeatedly peek and stop on a favorable result, call underpowered noise a
20	winner, or generalize beyond the tested population.
21
22