hypothesys
Skill1 starsChanged 13 days agoMITView on GitHub
Grill the user relentlessly about a data science or ML hypothesis until it is clear, falsifiable and tied to goal metrics: premises, baseline and success criteria. Use when starting an experiment, when the user wants to stress-test a hypothesis or idea, or says 'grill me', 'challenge my hypothesis' or 'is this testable'.
In this repository15 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- critic-reviewSkill
- developing-data-science-notebooksSkill
- expectation-driven-scienceSkill
- experiment-graphSkill
- experiment-pipelineSkill
- exp-reviewSkill
- let-it-ripSkill
- literature-researchSkill
- mad-science-setupSkill
- notify-gchat-human-waitSkill
- to-articleSkill
- to-docSkill
- what-i-doSkill
The file
Files
Interview the user relentlessly until you reach a shared understanding. This means making clear the hypothesis and premises, and the goal metrics to evaluate. Map this as a design tree: every decision branches into the decisions that hang off it.
Work in rounds like an interview. In the first round, the questions will be about things that are clear in the context; do not guess things you haven't heard yet. Ask every frontier question you think is necessary and give your recommended answer. Then wait for the user's answer for the next round.
❓ **Q1** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>
➡️ <your recommended answer>
---
❓ **Q2** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>
➡️ <your recommended answer>Each round, the user's answers reshape the tree: settled decisions push the frontier outward and unblock questions that depend on them. Recompute the frontier and ask the next round. A question whose answer depends on another question still open in this round belongs to a later round, not this one.
Finding facts is your job, never the user's. When a frontier question needs a fact from the environment (filesystem, tools, etc.), dispatch a sub-agent to find it; don't ask the user for anything you could look up yourself. Don't block on it: a running exploration is an unsettled prerequisite, so only the questions downstream of it wait for the sub-agent to report; ask the rest of the frontier now. The decisions are the user's: put each to them and wait. The session is done when the frontier is empty: every branch of the design tree is visited, and nothing is left silently assumed. Present the hypothesis, premises, and goal metrics reached at the end of the interview. Do not act on it until the user confirms you have reached a shared understanding.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.