Hypothesis Design
Forces a 5-part testable hypothesis statement instead of a vague A/B test guess
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 13, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 64cdce6
Ran a signup-form-shortening experiment through it: baseline produced a one-line guess, the skill forced a guardrail metric (activation rate) and a sample-size estimate the baseline skipped entirely.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Hypothesis Design does
Turns a fuzzy product problem into a falsifiable hypothesis with action, outcome, direction, users, and conditions, plus a measurement plan and narrative check. Use when scoping an A/B test or experiment before building it, not when analyzing results already collected.
How to install Hypothesis Design
git clone https://github.com/itseffi/agentic-os
cd agentic-os
mkdir -p ~/.claude/skills
cp -r .agents/skills/hypothesis-design ~/.claude/skills/hypothesis-design
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Hypothesis Design
-
/hypothesis-designForces a 5-part testable hypothesis statement instead of a vague A/B test guess
It also activates on plain-language prompts like these:
-
Help me write a proper five-part hypothesis before we build this A/B test -
I have a vague product problem, turn it into a falsifiable test hypothesis -
Before scoping this experiment, define outcome, direction, users, conditions
Frequently asked questions
- Is the Hypothesis Design skill free?
- Yes. The skill itself is free from itseffi/agentic-os. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Hypothesis Design work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 13, 2026. Verdict: Tested · Works. Ran a signup-form-shortening experiment through it: baseline produced a one-line guess, the skill forced a guardrail metric (activation rate) and a sample-size estimate the baseline skipped entirely.
- What is the Hypothesis Design SkillProof Score?
- 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Hypothesis Design?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Hypothesis Design with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.