agentleFS
fzfcleefzfclee/consulting-skills/skills/hypothesis-tree/SKILL.md

hypothesis-tree

Skill6 starsChanged 4 months agoApache-2.0View on GitHub

Use when analysis should proceed by testing likely explanations rather than describing the whole situation. Use when applying the Hypothesis Tree consulting method and when a user asks for Hypothesis Tree, method execution, structured diagnosis, or action planning.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill to run Hypothesis Tree as a practical consulting method, not as a generic framework explanation.

    line 9
  2. Very specific:

    If an input is missing, do not block automatically.

    line 25
  3. Distinctive:

    Do not use without a decision question and branches that can be tested with evidence.

    line 29
  4. Distinctive:

    Keep wording professional and plain enough that a smart non-specialist can use the output directly.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Hypothesis Tree

Use this skill to run Hypothesis Tree as a practical consulting method, not as a generic framework explanation.

Method Notes

  • Each branch must be testable.
  • Define disconfirmation evidence, not only supporting evidence.

Required Inputs

Collect or infer these inputs before execution:

  • core question
  • candidate explanations
  • available evidence
  • decision deadline

If an input is missing, do not block automatically. Mark it as missing, state the assumption used, and add a validation action.

When Not To Use

Do not use without a decision question and branches that can be tested with evidence. For open-ended exploration before hypotheses are available, use a mind map or affinity diagram.

Adjacent Methods

  • issue-tree: decompose the question into analysis branches before proposing explanations.
  • abductive-reasoning: rank the best current explanation from incomplete evidence.

Step-by-Step Execution

StepRequired inputHow to executeOutput
Frame hypothesis areaProblem, observed outcome, decision need.State what explanation or bet needs testing.Hypothesis frame.
Create competing hypothesesKnown facts and plausible causes.List alternative explanations or strategic bets.Top-level hypotheses.
Break into sub-hypothesesTop-level hypotheses.Make each branch testable and specific.Hypothesis tree.
Define testsHypothesis tree and available evidence.Specify supporting evidence and disconfirming evidence for each leaf.Test plan.
Prioritize testsTest plan, impact, speed, uncertainty.Choose fastest tests that could change the decision.Testing sequence.

Output Template

### 1. Decision Question
Question:
Scope:
Current evidence:
Decision deadline:

### 2. Competing Hypotheses
| Hypothesis | Sub-hypotheses | Evidence for | Evidence against | Confidence |
|---|---|---|---|---|
|  |  |  |  |  |

### 3. Disconfirming Tests
| Hypothesis | Cheapest decisive test | Owner | Timing | Threshold |
|---|---|---|---|---|
|  |  |  |  |  |

### 4. Priority
First hypothesis to test:
Why:
Decision unlocked:

### Evidence And Next Decision
- Confirmed facts:
- Assumptions:
- Missing evidence:
- Next action, owner, and timing:
- Expected signal and decision threshold:

Quality Gate

  • Produce the method-specific outputs for Decision Question, Competing Hypotheses, Disconfirming Tests; do not substitute a generic framework summary.
  • Tie every material score, causal claim, or stakeholder judgment to evidence or label it as an assumption.
  • Include at least one disconfirming check, sensitivity, alternative explanation, or failure condition appropriate to the method.
  • Convert the result into a named action or decision with an owner or stakeholder, timing, and observable signal.
  • State missing inputs and the smallest validation action instead of inventing precision.
  • Keep wording professional and plain enough that a smart non-specialist can use the output directly.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.