hypothesis-generation

Structured hypothesis formulation from observations. Use when you have experimental observations or data and need to formulate testable hypotheses with predictions, propose mechanisms, and design experiments to test them. Follows scientific method framework. For open-ended ideation use scientific-brainstorming; for automated LLM-driven hypothesis testing on datasets use hypogenic.

1.67x

Quality

67%

Does it follow best practices?

Impact

99%

1.67x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/hypothesis-generation/SKILL.md

Quality

Discovery

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This is a strong description that clearly articulates specific capabilities (hypothesis formulation, mechanism proposal, experiment design), provides explicit trigger conditions, and proactively disambiguates from related skills. The only minor note is the use of second person ('you have') which technically violates the third-person voice guideline, but the description is otherwise exemplary in all dimensions.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: 'formulate testable hypotheses with predictions', 'propose mechanisms', and 'design experiments to test them'. These are clear, actionable capabilities.	3 / 3
Completeness	Clearly answers both 'what' (formulate testable hypotheses, propose mechanisms, design experiments) and 'when' ('Use when you have experimental observations or data and need to formulate testable hypotheses'). Also includes explicit disambiguation guidance for related skills (scientific-brainstorming, hypogenic).	3 / 3
Trigger Term Quality	Includes strong natural keywords users would say: 'experimental observations', 'data', 'testable hypotheses', 'predictions', 'mechanisms', 'design experiments', 'scientific method'. These are terms a researcher or scientist would naturally use.	3 / 3
Distinctiveness Conflict Risk	Explicitly distinguishes itself from related skills ('For open-ended ideation use scientific-brainstorming; for automated LLM-driven hypothesis testing on datasets use hypogenic'), creating a clear niche for structured hypothesis formulation from observations. This anti-overlap guidance is excellent.	3 / 3
	Total	12 / 12 Passed

Implementation

35%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill is well-structured in terms of workflow sequencing and resource organization, but suffers significantly from verbosity and repetition. The LaTeX formatting instructions dominate the document and are repeated multiple times (overflow prevention appears 3+ times). The scientific methodology content is largely things Claude already knows, and the skill would benefit greatly from moving detailed formatting guidance to referenced files and focusing the main document on what's truly unique or non-obvious.

Suggestions

Cut the LaTeX formatting section by 70%+ — move detailed box usage, overflow prevention, and citation requirements entirely to FORMATTING_GUIDE.md and reference it with a single line. The current content repeats overflow prevention advice at least 3 times.

Remove or drastically shorten sections explaining concepts Claude already knows: what testability/falsifiability/parsimony mean, what literature synthesis involves, basic experimental design elements. Replace with a brief checklist or reference to the external criteria file.

Add a concrete worked example showing a real observation → hypothesis → prediction → experiment chain (even abbreviated) to make the workflow actionable rather than abstract.

Add explicit validation checkpoints in the workflow, e.g., 'Before proceeding to Step 4, verify you have identified at least 2 conflicting findings or gaps' and 'After Step 5, discard any hypothesis scoring below threshold on testability/falsifiability.'

Dimension	Reasoning	Score
Conciseness	Extremely verbose at ~250+ lines. Massive amounts of content Claude already knows (how to structure hypotheses, what testability means, what falsifiability is). The LaTeX formatting instructions are repeated multiple times (page overflow prevention is stated 3+ times with nearly identical advice). The 'When to Use This Skill' section lists obvious use cases. Much of this could be cut by 60-70%.	1 / 3
Actionability	Provides some concrete guidance like the xelatex compilation commands and the bash command for schematics, but most of the workflow steps are abstract descriptions of what to do rather than executable instructions. References external files (hypothesis_quality_criteria.md, experimental_design_patterns.md, templates) which is good, but the inline content is largely descriptive rather than providing concrete examples of actual hypothesis formulation.	2 / 3
Workflow Clarity	The 8-step workflow is clearly sequenced and numbered, which is good. However, there are no validation checkpoints or feedback loops — no step says 'verify your hypotheses meet criteria X before proceeding' or 'if literature search yields insufficient results, broaden scope.' The LaTeX compilation steps are clear but the scientific workflow itself lacks explicit go/no-go decision points.	2 / 3
Progressive Disclosure	References external files appropriately (references/, assets/, related skills) and has a clear Resources section. However, the main SKILL.md is monolithic — the extensive LaTeX formatting instructions (page overflow prevention, box usage, citation requirements) should be in the FORMATTING_GUIDE.md rather than duplicated inline. The schematics section is also overly detailed for what should be a cross-reference.	2 / 3
	Total	7 / 12 Passed

Validation

90%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	10 / 11 Passed

Repository: K-Dense-AI/claude-scientific-skills
Commit: 25e1c0f

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.