- Home
- Skills
- Sammcj
- Agentic Coding
- Critical Thinking Logical Reasoning
critical-thinking-logical-reasoning_skill
- Python
110
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill sammcj/agentic-coding --skill critical-thinking-logical-reasoning- SKILL.md3.5 KB
Overview
This skill performs structured critical thinking and logical reasoning analysis on written content when explicitly requested. It focuses on clarifying claims, exposing weak evidence or hidden assumptions, and producing concise, actionable critiques that preserve signal and domain insight. The output is practical, decision-focused and suitable for reports, meetings or policy review.
How this skill works
I first restate the core argument in a way the author would recognise, ensuring accurate representation before critique. I then identify core claims, evaluate supporting evidence, detect logical issues and surface hidden assumptions. The response follows a fixed structure: Summary, Key Issues, Questions to Probe and Bottom Line, emphasising what matters for decisions. Recommendations highlight how to strengthen fixable flaws while noting where uncertainty or missing context limits judgement.
When to use it
- Analysing articles, opinion pieces or policy briefs for logical soundness
- Reviewing meeting notes, transcripts or summaries to preserve signal quality
- Preparing briefings where decisions depend on argument strength
- Assessing persuasive messaging for fallacies or overstated claims
- Before adopting recommendations from reports or white papers
Best practices
- Start by restating the argument in the author’s words to avoid straw-manning
- Separate core claims from supporting points before judging evidence
- Prioritise issues that materially affect the conclusion over minor errors
- Ask targeted probe questions that reveal hidden assumptions or missing evidence
- Offer concrete fixes for weaknesses rather than only identifying faults
- Be charitable: assume good intentions and label weak reasoning as 'flawed' not 'wrong'
Example use cases
- A policymaker needs a short critique of a white paper before a vote
- A product team wants to check whether a market-analysis blog overstates conclusions
- A meeting chair wants a high-signal summary of a long transcript for post-meeting action items
- A communications lead needs to remove persuasive fallacies from draft messaging
- An executive requests quick checks of consultant recommendations for hidden dependencies
FAQ
I focus on reasoning and internal consistency; I do not perform comprehensive fact-checking unless a claim is obviously implausible or self-contradictory.
Will you preserve tone and intent?
Yes. I begin by restating the argument in terms the author would accept to avoid misrepresentation and preserve intent.
Do you handle code or technical models?
No. This skill is for written arguments, reports, articles and transcripts, not for analysing source code or technical model implementations.