counterargument
Generates rigorous devil's advocate critiques from methodological, theoretical, and practical perspectives. Use when the user asks to challenge their own work, find weaknesses, stress-test assumptions, anticipate reviewer objections, or argue against their research claims. Run BE
What it does
Counterargument Generation
Generate adversarial critiques to stress-test research claims before writing or submitting.
Perspectives
| Perspective | Icon | Focus |
|---|---|---|
| Methodological | ๐ฌ | Study design, measurement validity, statistical approach, reproducibility |
| Theoretical | ๐ | Conceptual framework, alternative explanations, literature gaps |
| Practical | โ๏ธ | Feasibility, generalizability, real-world applicability, ethical concerns |
Workflow
- Identify the core claims โ extract the main hypotheses or arguments
- Generate 4โ6 critiques โ at least one per perspective, more for weaker claims
- Rate severity โ HIGH (fatal flaw), MEDIUM (significant concern), LOW (minor point)
- Provide rebuttal strategy โ for each critique, suggest how to respond or strengthen the research
Output Format
# โ๏ธ Counterargument Analysis
## Summary
[2โ3 sentences: overall assessment of the research's vulnerability to critique]
## Counterarguments
### ๐ Theoretical Critique โ [SHORT TITLE]
**Challenged Claim:** [Exact claim being challenged]
**Critique:** [Specific, evidence-grounded critique. Reference timescales, measurement limits,
alternative frameworks, or published contradictory evidence where possible.]
**Rebuttal Strategy:** [Concrete suggestion: what experiment, analysis, or argument would
address this critique]
---
### ๐ฌ Methodological Critique โ [SHORT TITLE]
...
Quality criteria
- Specific: Reference exact claims, timescales, measurements โ not vague "this is uncertain"
- Grounded: Connect to actual limitations in the content, not generic criticism
- Expert-level: Write as a senior reviewer at a top journal would
- Constructive: Every critique ends with a path forward (rebuttal strategy)
Examples of strong vs weak critiques
Weak: "The sample size might be too small."
Strong: "The proposed sample of n=15 animals achieves 80% power only for effect sizes of ฮทยฒโฅ0.25. Given the typical effect sizes in STDP studies (ฮทยฒ=0.10โ0.15, e.g., Bi & Poo 1998), this is underpowered. A power analysis targeting ฮทยฒ=0.12 would require n=38 per condition."
Capabilities
Install
Quality
deterministic score 0.47 from registry signals: ยท indexed on github topic:agent-skills ยท 37 github stars ยท SKILL.md body (2,206 chars)