Hypothesis Generation
Format structured scientific hypothesis reports with a reusable LaTeX template and color-coded sections for solo researchers documenting phenomena.
Overview
Hypothesis Generation is an agent skill for the Idea phase that produces a LaTeX-structured scientific hypothesis report with color-coded hypotheses, predictions, and evidence sections.
Install
npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill hypothesis-generationWhat is this skill?
- XeLaTeX/LuaLaTeX-ready hypothesis_generation style package with executive summary, prediction, and limitation box enviro
- Five distinct hypothesis color tokens plus utility colors for predictions, evidence, comparisons, and limitations
- natbib-ready article scaffold for citable hypothesis-generation reports
- Quick-reference formatting guide paired with full SKILL.md documentation
- Standard compile loop: xelatex → bibtex → xelatex ×2
- 5 hypothesis color slots
- 4 utility color roles for predictions evidence comparisons limitations
Adoption & trust: 591 installs on skills.sh; 27.6k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have a research phenomenon and informal notes but no consistent, citable document that lays out competing hypotheses and testable predictions.
Who is it for?
Indie researchers or technical solo builders documenting multi-hypothesis explanations before designing experiments or validation prototypes.
Skip if: Teams that only need a plain Markdown brainstorm with no PDF deliverable or who do not use LaTeX in their toolchain.
When should I use this skill?
You need a formal hypothesis-generation PDF with executive summary, numbered hypotheses, and bibliography support from agent-assisted research notes.
What do I get? / Deliverables
After the skill runs, you have a compilable LaTeX report template filled with structured hypothesis blocks ready for review or bibliography integration.
- Compilable .tex hypothesis report
- PDF-ready structured sections for hypotheses and predictions
Recommended Skills
Journey fit
Hypothesis generation is the earliest structured research artifact—before validation experiments—so it shelves under Idea phase research. Research subphase covers framing phenomena, competing explanations, and testable predictions before prototyping or data collection.
How it compares
Use instead of generic article templates when you need hypothesis-specific boxes and color semantics rather than a one-size-fits-all paper layout.
Common Questions / FAQ
Who is hypothesis-generation for?
Solo builders, student researchers, and indie scientists who want agent help turning qualitative research ideas into a formal LaTeX hypothesis report.
When should I use hypothesis-generation?
Use it in the Idea research phase when framing a phenomenon, and optionally revisit in Validate scope when tightening predictions before experiments.
Is hypothesis-generation safe to install?
Review the Security Audits panel on this Prism page for install risk and file integrity before adding the skill to your agent.
SKILL.md
READMESKILL.md - Hypothesis Generation
# Hypothesis Generation Report - Formatting Quick Reference ## Overview This guide provides quick reference for using the hypothesis generation LaTeX template and style package. For complete documentation, see `SKILL.md`. ## Quick Start ```latex % !TEX program = xelatex \documentclass[11pt,letterpaper]{article} \usepackage{hypothesis_generation} \usepackage{natbib} \title{Your Phenomenon Name} \begin{document} \maketitle % Your content \end{document} ``` **Compilation:** Use XeLaTeX or LuaLaTeX for best results ```bash xelatex your_document.tex bibtex your_document xelatex your_document.tex xelatex your_document.tex ``` ## Color Scheme Reference ### Hypothesis Colors - **Hypothesis 1**: Deep Blue (RGB: 0, 102, 153) - Use for first hypothesis - **Hypothesis 2**: Forest Green (RGB: 0, 128, 96) - Use for second hypothesis - **Hypothesis 3**: Royal Purple (RGB: 102, 51, 153) - Use for third hypothesis - **Hypothesis 4**: Teal (RGB: 0, 128, 128) - Use for fourth hypothesis (if needed) - **Hypothesis 5**: Burnt Orange (RGB: 204, 85, 0) - Use for fifth hypothesis (if needed) ### Utility Colors - **Predictions**: Amber (RGB: 255, 191, 0) - For testable predictions - **Evidence**: Light Blue (RGB: 102, 178, 204) - For supporting evidence - **Comparisons**: Steel Gray (RGB: 108, 117, 125) - For critical comparisons - **Limitations**: Coral Red (RGB: 220, 53, 69) - For limitations/challenges ## Custom Box Environments ### 1. Executive Summary Box ```latex \begin{summarybox}[Executive Summary] Content here \end{summarybox} ``` **Use for:** High-level overview at the beginning of the document --- ### 2. Hypothesis Boxes (5 variants) ```latex \begin{hypothesisbox1}[Hypothesis 1: Title] \textbf{Mechanistic Explanation:} [2-3 paragraphs explaining HOW and WHY] \textbf{Key Supporting Evidence:} \begin{itemize} \item Evidence point 1 \citep{ref1} \item Evidence point 2 \citep{ref2} \end{itemize} \textbf{Core Assumptions:} \begin{enumerate} \item Assumption 1 \item Assumption 2 \end{enumerate} \end{hypothesisbox1} ``` **Available boxes:** `hypothesisbox1`, `hypothesisbox2`, `hypothesisbox3`, `hypothesisbox4`, `hypothesisbox5` **Use for:** Presenting each competing hypothesis with its mechanism, evidence, and assumptions **Best practices for 4-page main text:** - Keep mechanistic explanations to 1-2 brief paragraphs only (6-10 sentences max) - Include 2-3 most essential evidence points with citations - List 1-2 most critical assumptions - Ensure each hypothesis is genuinely distinct - All detailed explanations go to Appendix A - **Use `\newpage` before each hypothesis box to prevent overflow** - Each complete hypothesis box should be ≤0.6 pages --- ### 3. Prediction Box ```latex \begin{predictionbox}[Predictions: Hypothesis 1] \textbf{Prediction 1.1:} [Specific prediction] \begin{itemize} \item \textbf{Conditions:} When/where this applies \item \textbf{Expected Outcome:} Specific measurable result \item \textbf{Falsification:} What would disprove it \end{itemize} \end{predictionbox} ``` **Use for:** Testable predictions derived from each hypothesis **Best practices for 4-page main text:** - Make predictions specific and quantitative when possible - Clearly state conditions under which prediction should hold - Always specify falsification criteria - Include only 1-2 most critical predictions per hypothesis in main text - Additional predictions go to appendices --- ### 4. Evidence Box ```latex \begin{evidencebox}[Supporting Evidence] Content discussing supporting evidence \end{evidencebox} ``` **Use for:** Highlighting key supporting evidence or literature synthesis **Best practices:** - Use sparingly in main text (detailed evidence goes in Appendix A) - Include citations for all evidence - Focus on most compelling evidence --- ### 5. Comparison Box ```latex \begin{comparisonbox}[H1 vs. H2: Key Distinction] \textbf{Fundamental Difference:} [Description o