Skip to content
Skills Library
Science & research · Research

hypothesis-generation

/hypothesis-generation

Generate testable hypotheses from literature and observation. Use for converting research gaps, contradictory findings, cross-domain analogies, or clinical observations into falsifiable predictions for experimental design.

Licence
MIT
Last repository change
24.06.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
1
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 24.06.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What to check before installing
  • SKILL.md zawiera komentarze HTML (1). Nie widać ich w podglądzie Markdown, a model je czyta.

What it does and when to use it

Generate testable hypotheses from literature and observation. Use for converting research gaps, contradictory findings, cross-domain analogies, or clinical observations into falsifiable predictions for experimental design.

Original author description

Generate testable hypotheses from literature and observation. Use for converting research gaps, contradictory findings, cross-domain analogies, or clinical observations into falsifiable predictions for experimental design.

How it works

  1. Hard rules
  2. When to use
  3. When NOT to use
  4. Prerequisites
  5. Core workflow
  6. Code patterns
  7. Common pitfalls
  8. Validation
  9. References
  10. Changelog
Original SKILL.md headings · 326 lines, 1,430 words
SKILL.md preview
---

name: hypothesis-generation
description: "Generate testable hypotheses from literature and observation. Use for converting research gaps, contradictory findings, cross-domain analogies, or clinical observations into falsifiable predictions for experimental design."
license: MIT
---
<!-- metadata:
  category: scientific-thinking
  oversight: very-high
  version: 1.0.0
  author: Pradyumna Jayaram
  tags: [scientific-thinking, research]
  difficulty: intermediate
-->

# Hypothesis Generation

## Hard rules

1. **No fabricated citations.** Every cited work must resolve to a verifiable
2. **No claim without provenance.** Every quantitative or factual claim
3. **No silent failure.** Every script invocation, API call, or tool use must declare its exit status and what to do on non-zero. A skill that silently swallows errors is a violation.

## When to use

- Identifying testable predictions from literature gaps or contradictions
- Converting brainstorming ideas into experimentally tractable questions
- Formulating hypotheses from clinical or observational patterns
- Developing grant-specific aims with clear predictions
- Designing experiments to distinguish between competing theories
- Applying cross-domain analogies to generate novel predictions
- Moving from exploratory analysis to confirmatory testing

## When NOT to use

- Early-stage exploration without clear direction (use brainstorming)
- Evaluating evidence quality of existing hypotheses (use critical thinking)
- Confirming established knowledge (that's verification, not hypothesis generation)
- When a phenomenon is already well-explained and no gap exists
- For purely descriptive research without predictive component

## Prerequisites

- Sufficient domain knowledge to understand the current state of research
- Familiarity with relevant literature and open questions
- Basic understanding of experimental design principles
- Access to data or observations to build upon
- Awareness of competing theories or explanations in the field

## Core workflow

### 1. Identify the hypothesis source

Determine where the hypothesis idea originates:

**Knowledge gaps in literature**
- What questions remain unanswered?
- What has been overlooked or insufficiently studied?
- What contradictions exist between studies?

**Contradictory findings**
- Studies that report opposing results
- Inconsistent effect sizes across populations
- Conflicting mechanistic explanations

**Cross-domain analogies**
- Insights from other fields that might apply
- Biological parallels to engineering solutions
- Methodological advances from adjacent areas

**Methodological advances**
- New techniques that enable previously impossible tests
- More precise measurements or manipulations
- Scale or resolution improvements

**Clinical or observational patterns**
- Unexpected observations in practice
- Patient/subject patterns noticed but unexplained
- Natural experiments or quasi-experiments

**Theoretical predictions**
- Predictions from models that need testing
- Implications of established theories
- Boundary conditions not yet explored

### 2. Specify the hypothesis anatomy

A well-formed hypothesis has five components:

**1. Variables**
- Independent variable (what you manipulate or vary)
- Dependent variable (what you measure)
- Control variables (what you hold constant)

**2. Directional prediction**
- What you expect to happen
- Specific direction (increase/decrease, positive/negative)
- Magnitude or effect size expectations

**3. Mechanism (if causal)**
- Why you expect this to happen
- Underlying causal chain
- Theoretical basis for the prediction

**4. Scope/conditions**
- When the hypothesis applies
- Boundary conditions
- Populations, settings, or contexts

**5. Operationalization**
- How each variable will be measured
- Cutoffs or thresholds for categorization
- Specific experimental protocols

**Template:**"
> "We hypothesize that [IV] will [directionally affect] [DV] in [population/samples], because [mechanism], and this will be measured by [operationalization]."

### 3. Apply the falsifiability test

Following Popper's criterion, a hypothesis must be falsifiable:

**Questions to ask:**
- Could the hypothesis be proven wrong by observation?
- Is there a conceivable result that would contradict it?
- Are the variables operacionais in a way that allows rejection?

**Types of unfalsifiable hypotheses to avoid:**
- No clear prediction (too vague)
- Invulnerable to disconfirmation (always interprets结果是"支持")
- Tautological (true by definition)
- Untestable with available methods

**Making hypotheses falsifiable:**
- Specify exact predictions, not just direction
- State magnitude expectations
- Define rejection criteria in advance
- Identify alternative explanations

### 4. Design a test

How would you test this hypothesis?

**Requirements:**
- An experiment or observation that could yield conflicting results
- A way to measure the dependent variable
- Appropriate controls
- Sufficient sample/power

**Design considerations:**
- What design can differentiate this from alternatives?
- What would confirm? What would falsify?
- What are plausible confounds?
- How will you handle ambiguity in results?

### 5. Specify alternatives and boundary conditions

A strong hypothesis acknowledges:

**Competing hypotheses**
- What else could explain the results?
- What are the leading alternatives?
- How does this hypothesis differ?

**Boundary conditions**
- When would you expect this not to hold?
- What are the limits of the prediction?
- What contextual factors matter?

**Effect size expectations**
- What magnitude of effect makes the hypothesis "true"?
- What magnitude suggests falsification?
- Is the effect clinically/practically significant?

## Code patterns

### Hypothesis specification template

```
HYPOTHESIS: [Number]

Variables:
- Independent variable: [precise definition]
- Dependent variable: [precise definition]
- Control variables: [list]

Prediction:
- We expect [IV] to [increase/decrease] [DV] by approximately [magnitude/percentage].

Mechanism:
- [IV] affects [DV] through [mechanism], based on [theory/prior work].

Scope:
- This prediction applies to [population/context].
- We expect this in [conditions], but not when [conditions].

Operationalization:
- [IV] will be operationalized as: [specific measurement]
- [DV] will be operationalized as: [specific measurement]

Rejection criteria:
- If [observed result], we reject the hypothesis.

Alternative explanations:
1. [Competing hypothesis 1]
2. [Competing hypothesis 2]
```

### Example: From gap to hypothesis

**Source:** Two studies on mitochondrial function in aging report conflicting results - one shows decline, one shows no change.

**Gap:** Why the inconsistency? One uses tissue homogenates, one uses single cells.

**Hypothesis Generation:**

```
HYPOTHESIS 1: Tissue-level measurements obscure cell-type heterogeneity
in mitochondrial function decline during aging.

Variables:
- Independent variable: Measurement scale (tissue homogenate vs. single cell)
- Dependent variable: Mitochondrial function (ATP production rate)
- Control variables: Age, species, tissue type

Prediction: Single-cell measurements will reveal significant
age-related decline that tissue homogenates obscure by averaging
across cell subpopulations with different function trajectories.

Mechanism: Cell-to-cell variation increases with age; averaging
masks decline in high-function cells by diluting with low-function cells.

Scope: Applicable to post-mitotic tissues (neurons, muscle).
Not applicable to proliferating tissues with ongoing stem cell input.

Rejection criteria: If single-cell measurements show no greater
age-related decline than tissue homogenates, reject hypothesis.
```

### Contradictory findings resolution framework

```
CONFLICT RESOLUTION: [Citation A] vs. [Citation B]

Both studies claim:
- Study A: [Claim]
- Study B: [Opposing claim]

Potential explanations:
1. Population differences (species, age, sex)
2. Methodological differences (assay, timing)
3. Context differences (in vivo vs. in vitro)
4. Statistical artifacts (power, analysis)

HYPOTHESIS to resolve:
We hypothesize that [variable X] explains the discrepancy,
specifically [prediction].
```

### From clinical observation to hypothesis

```
CLINICAL OBSERVATION:
[Describe unexpected pattern noticed in practice]

PATTERN TO EXPLAIN:
[What was observed]

LEADING HYPOTHESES:
1. [Potential explanation 1]
2. [Potential explanation 2]

HYPOTHESIS TO TEST:
We hypothesize that [specific mechanism], based on
[observations from basic science / analogous conditions].

Testable prediction:
[What would we expect to see if this is correct?]
```

## Common pitfalls

- **Vague predictions:** "X affects Y" is not a hypothesis; "X increases Y by 20-30%" is.
- **Unfalsifiable wording:** Avoid "may," "could," or "might" when making predictions.
- **Mechanical application:** Don't force every observation into hypothesis form when it's better described as an exploratory finding.
- **Missing mechanism:** Without a plausible mechanism, predictions are arbitrary.
- **Ignoring alternatives:** Strong hypotheses specify what's being distinguished from.
- **Unrealistic scope:** Start with narrowly testable hypotheses before combining.
- **Shoehorning:** Don't try to find evidence FOR your hypothesis; design tests that could falsify it.
- **Neglecting boundary conditions:** Specify when your hypothesis should NOT hold.

## Validation

How to know your hypothesis generation was successful:

- The hypothesis is stated as a specific, testable prediction
- Variables are clearly defined and operacionalized
- The hypothesis could be falsified by a plausible outcome
- There's a clear experimental design to test it
- Alternative explanations are acknowledged
- The scope and boundary conditions are specified
- The mechanism or theoretical basis is provided
- Effect size expectations are stated
- The hypothesis addresses a genuine gap or contradiction

## References

- **Related ors-* skills:**
  - ors-scientific-thinking-brainstorming (for initial exploration)
  - ors-scientific-thinking-critical-thinking (for evaluation)
  - ors-scientific-thinking-perspective-tour (for multi-perspective framing)
  - ors-scientific-thinking-failure-handling (if results are negative)
  - ors-research-grants-specific-aims (for hypothesis in grant context)

- **External resources:**
  - Popper, K.R.. The Logic of Scientific Discovery
  - Lakatos, I.. Falsification and the Methodology of Scientific Research Programmes
  - CONSORT Statement - hypothesis reporting in trials
  - STROBE Statement - hypothesis reporting in observational studies

## Changelog

- 1.0.0 (2026-06-10): Initial adaptation by Pradyumna Jayaram, integrating Popper's falsifiability criterion, Lakatos's research programmes, and structured frameworks for converting knowledge gaps and contradictions into testable hypotheses.

## Cross-references

Other skills in this category:

- brainstorming
- critical-thinking
- failure-handling
- perspective-tour

Versions

RepositoryScoreLicence
This page · pradyumnasagar/open-research-skills100MIT
foryourhealth111-pixel/Vibe-Skills100Apache-2.0
xinhi6716/axiom-atropos100MIT
raoofaltaher/science-ultrapowers100MIT
PracticalSwan/agent-skills100MIT
K-Dense-AI/scientific-agent-skills100MIT
adeerkhan/vitruvius100MIT
lamm-mit/scienceclaw100Apache-2.0
aipoch/medical-research-skills100MIT

Database snapshot: 13.11.2026 · package version 1e4b52d56d73 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 1 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100hypothesis
    Generates falsifiable, testable research hypotheses from notes, documents, and research content. Use when the user asks to brainstorm hypotheses, generate research questions, identify testable predictions, or discover patterns across their notes. Do NOT use for general Q&A —…
  • 100alterlab-hypothesis-gen
    Formulates structured, testable hypotheses from experimental observations using a scientific-method framework — derives predictions, proposes mechanisms, and designs experiments to test them, then renders a LaTeX report. Use when turning observations or data into falsifiable,…
  • 100hypothesis-gen
    Structured hypothesis generation workflow. Use when: user needs to formulate testable scientific hypotheses from observations, gaps, or preliminary data. NOT for: testing hypotheses or running experiments.
  • 100hypothesis-testing
    You must use this when formulating testable hypotheses, designing experimental controls, or defining falsification criteria.
  • 91scholar-lit-review-hypothesis
    Conduct a systematic literature review AND develop theory and hypotheses in one integrated workflow. Use when the user wants to map existing knowledge, identify gaps, select theoretical frameworks that address those gaps, specify mechanisms, and derive testable hypotheses —…
  • 90scienceclaw-generation
    Generate scientific hypotheses, experimental designs, and paper drafts. Use when: user asks to propose hypotheses, design experiments, or write scientific content. NOT for: data analysis or literature search.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.