Hypothesis Generation

When you've got experimental data or observations and need to develop testable scientific hypotheses, this gives you a structured workflow. It walks you through literature search (PubMed and web), generating 3-5 competing mechanistic explanations, evaluating them against quality criteria like falsifiability and parsimony, then designing experiments to test each one. The output is a LaTeX report with hypothesis boxes, predictions, and mandatory AI-generated schematics using the scientific-schematics skill. It's opinionated about structure and leans heavy on visual documentation, which makes sense for research work. Best for actual scientific inquiry where you need rigorous hypothesis formulation, not casual brainstorming.

skills.sh installs: 2.2K
GitHub stars: 56.6K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add k-dense-ai/scientific-agent-skills --skill hypothesis-generation --agent claude-code

Installs into .claude/skills of the current project.

Files

Scientific Hypothesis Generation

Turn an observation into a transparent set of candidate explanations and tests. A hypothesis is a proposal to be challenged, not a finding, fact, diagnosis, or recommendation.

Non-negotiable boundaries

Before using unpublished, sensitive, controlled, personal, proprietary, export-controlled, or security-relevant material:

Confirm authorization and the applicable institutional, funder, publisher, data-use, privacy, and AI policies.
Keep the material local unless an authorized human explicitly approves a named external destination and data scope.
Minimize inputs. Do not place sensitive or unpublished data in web searches or external AI systems without authorization.
Stop at the appropriate human, animal, biosafety, dual-use, data-governance, or regulatory gate.

Never:

SKILL.md on GitHub