Applies rigorous critical analysis to evaluate claims, arguments, and research. Use when evaluating evidence quality, peer reviewing content, assessing argument validity, or identifying weaknesses in reasoning. Triggers on phrases like "critically analyze", "evaluate this", "review this paper", "check the logic", "assess the evidence", "find flaws", "peer review".
View on GitHubpoemswe/co-researcher
co-researcher
January 24, 2026
Select agents to install to:
npx add-skill https://github.com/poemswe/co-researcher/blob/main/skills/critical-analysis/SKILL.md -a claude-code --skill critical-analysisInstallation paths:
.claude/skills/critical-analysis/# Critical Analysis Framework
This skill guides rigorous critical evaluation of claims, arguments, and research.
## Phase 1: Content Mapping
### Claim Extraction
Identify all claims in the material:
1. **Central claim**: The main argument or thesis
2. **Supporting claims**: Claims used to support the central claim
3. **Implicit claims**: Unstated assumptions
4. **Hedged claims**: Qualified or conditional statements
### Argument Structure Mapping
```
Conclusion (Central Claim)
↑
Premise 1 + Premise 2 + Premise 3
↑ ↑ ↑
[Evidence] [Evidence] [Evidence]
```
### Stakeholder Context
- Who created this content?
- What are their credentials?
- What are potential motivations/interests?
- Who funded the work?
**CHECKPOINT**: Confirm scope of analysis with user.
## Phase 2: Evidence Assessment
### Evidence Inventory
| Claim | Evidence Provided | Evidence Type | Quality |
|-------|-------------------|---------------|---------|
| [Claim] | [What evidence] | [Type] | [Rating] |
### Evidence Types Hierarchy
(Strongest to weakest)
1. Systematic reviews/meta-analyses
2. Randomized controlled trials
3. Cohort studies
4. Case-control studies
5. Cross-sectional studies
6. Case reports
7. Expert opinion
8. Anecdote
### Evidence Quality Markers
**Strong evidence**:
- Peer-reviewed
- Replicable methodology
- Adequate sample size
- Appropriate controls
- Transparent reporting
**Weak evidence**:
- Not peer-reviewed
- Vague methodology
- Small sample
- No controls
- Selective reporting
## Phase 3: Logical Analysis
### Deductive Validity Check
For deductive arguments:
- Are premises true?
- Does conclusion follow necessarily from premises?
- Is the logical form valid?
### Inductive Strength Check
For inductive arguments:
- Is the sample representative?
- Is the sample large enough?
- Are there counterexamples?
- How strong is the correlation?
### Common Fallacy Scan
**Relevance Fallacies**:
- [ ] Ad hominem (attacking person, not argument)
- [ ]