How do I install the Stats skill?

Run npx skillmds@latest add richard-kim-79/stats in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Stats safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Stats?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Stats free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

richard-kim-79 (@richard-kim-79) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Stats by richard-kim-79 — Agent Skill

Name: Stats
Author: richard-kim-79

Stats

Detects statistical errors, logical fallacies, and methodological issues in research content. Checks for p-hacking, correlation/causation confusion, underpowered samples, multiple comparisons problems, overgeneralization, and other common fallacies. Use when the user asks to validate statistics, audit quantitative claims, check methodology, or find logical errors. Returns minimal output on purely theoretical content — most useful after empirical data is present.

File contents

Statistical Validator

Detect statistical errors and methodological fallacies in research content.

Fallacy Types Detected

Type	Description
`P_HACKING`	Selective reporting, post-hoc hypothesis changes, stopped when p<0.05
`CORRELATION_CAUSATION`	Causal claims from correlational data
`SMALL_SAMPLE`	Sample size insufficient for claimed effect size
`MULTIPLE_COMPARISONS`	Multiple tests without Bonferroni/FDR correction
`OVERGENERALIZATION`	Results from specific sample applied to broader population
`CIRCULAR_REASONING`	Conclusion assumes what it claims to prove
`CHERRY_PICKING`	Selective evidence presentation
`EFFECT_SIZE_MISSING`	Statistical significance without practical effect size
`CONFOUND`	Alternative explanations not controlled for

Severity Levels

HIGH — fatal flaw; invalidates the finding
MEDIUM — significant concern; finding is weakened
LOW — minor issue; addressable in discussion

Output Format

When issues are found:

# 📐 Statistical Validation

> Found **3 issue(s)** requiring attention.

## Issues

### 🔴 HIGH — P_HACKING

**Post:** [reference to source]
**Claim:** "[exact statistical claim]"
**Issue:** [specific explanation of the problem]
**Suggestion:** [concrete fix]

---

### 🟡 MEDIUM — CORRELATION_CAUSATION
...

## Summary
[Overall assessment + priority order for fixes]

When no issues are found:

# 📐 Statistical Validation

> ✅ No statistical issues detected.

## Assessment
[Explanation: e.g., "This content is theoretical/conceptual and contains no quantitative claims to validate."]

## Proactive Checklist
When empirical data is added, watch for:
- [ ] [Domain-specific statistical concern 1]
- [ ] [Domain-specific statistical concern 2]

When content has no statistics

If the content is purely theoretical or conceptual, note this explicitly and provide a domain-appropriate proactive checklist. Do NOT generate phantom issues.

---name: statsdescription: Detects statistical errors, logical fallacies, and methodological issues in research content. Checks for p-hacking, correlation/causation confusion, underpowered samples, multiple comparisons problems, overgeneralization, and other common fallacies. Use when the user asks to validate statistics, audit quantitative claims, check methodology, or find logical errors. Returns minimal output on purely theoretical content — most useful after empirical data is present.license: MIT---# Statistical ValidatorDetect statistical errors and methodological fallacies in research content.## Fallacy Types Detected| Type | Description ||------|-------------|| `P_HACKING` | Selective reporting, post-hoc hypothesis changes, stopped when p<0.05 || `CORRELATION_CAUSATION` | Causal claims from correlational data || `SMALL_SAMPLE` | Sample size insufficient for claimed effect size || `MULTIPLE_COMPARISONS` | Multiple tests without Bonferroni/FDR correction || `OVERGENERALIZATION` | Results from specific sample applied to broader population || `CIRCULAR_REASONING` | Conclusion assumes what it claims to prove || `CHERRY_PICKING` | Selective evidence presentation || `EFFECT_SIZE_MISSING` | Statistical significance without practical effect size || `CONFOUND` | Alternative explanations not controlled for |## Severity Levels- **HIGH** — fatal flaw; invalidates the finding- **MEDIUM** — significant concern; finding is weakened- **LOW** — minor issue; addressable in discussion## Output Format### When issues are found:```markdown# 📐 Statistical Validation> Found **3 issue(s)** requiring attention.## Issues### 🔴 HIGH — P_HACKING**Post:** [reference to source]**Claim:** "[exact statistical claim]"**Issue:** [specific explanation of the problem]**Suggestion:** [concrete fix]---### 🟡 MEDIUM — CORRELATION_CAUSATION...## Summary[Overall assessment + priority order for fixes]```### When no issues are found:```markdown# 📐 Statistical Validation> ✅ No statistical issues detected.## Assessment[Explanation: e.g., "This content is theoretical/conceptual and contains no quantitative claims to validate."]## Proactive ChecklistWhen empirical data is added, watch for:- [ ] [Domain-specific statistical concern 1]- [ ] [Domain-specific statistical concern 2]```## When content has no statisticsIf the content is purely theoretical or conceptual, note this explicitly and provide a domain-appropriate proactive checklist. Do NOT generate phantom issues.

Frequently asked questions