FerroxLabs Critical Thinker
When to Use
Process
Gather requirements. Ask the user clarifying questions about their specific context, goals, constraints, and experience level.
Analyze the situation. Review the information provided and identify key factors, challenges, and opportunities relevant to critical thinker.
Develop the framework. Create a structured approach tailored to the user's needs, incorporating best practices and domain-specific considerations.
Deliver actionable output. Present specific, implementable recommendations with clear rationale, timelines, and success criteria.
Address edge cases. Proactively identify potential issues, alternative approaches, and contingency plans.
Use this skill when:
- User needs guidance on critical thinker
- User asks about critical thinker best practices or techniques
- User wants a structured approach to critical thinker
Do NOT use this skill when:
- A more specialized skill exists for the specific subtopic
- The request is outside the scope of critical thinker
Questions to Ask First
Before evaluating any claim or argument, clarify:
- What claim is being made? (State it precisely and completely)
- Who is making the claim? (What are their credentials, incentives, and potential biases)
- What evidence is presented? (Data, anecdotes, expert opinion, logic)
- What is the context? (Media article, research paper, political speech, advertisement, conversation)
- What decision depends on this analysis? (What are you trying to decide)
- What is the cost of being wrong? (Stakes determine how rigorous you need to be)
- What do you already believe about this topic? (Your own biases to watch for)
Logical Fallacies Encyclopedia
Fallacies of Relevance
1. AD HOMINEM (Attacking the Person)
Attacking the person making the argument instead of the argument.
"You can't trust his climate research -- he drives an SUV."
TEST: Does the person's character actually invalidate the data?
2. APPEAL TO AUTHORITY (Argumentum ad Verecundiam)
Using an authority figure's opinion as proof, especially outside
their expertise.
"A famous actor says this supplement works, so it must."
TEST: Is the authority an expert in THIS specific field?
NOTE: Expert opinion in their field IS evidence, just not proof.
3. APPEAL TO EMOTION (Argumentum ad Passiones)
Using emotional manipulation instead of logical reasoning.
"Think of the children!" (without evidence of actual harm)
TEST: Remove the emotional framing. Is there a logical argument left?
4. APPEAL TO TRADITION (Argumentum ad Antiquitatem)
"We've always done it this way, so it must be right."
TEST: Is there evidence that the traditional way is effective,
or just that it's old?
5. APPEAL TO NATURE
"Natural is always better than artificial."
TEST: Arsenic is natural. Insulin is artificial. Is "natural"
actually relevant to whether something is good?
6. APPEAL TO POPULARITY (Argumentum ad Populum)
"Millions of people believe this, so it must be true."
TEST: Has popular opinion ever been wrong? (Yes, constantly.)
7. RED HERRING
Introducing an irrelevant topic to divert attention.
"Why worry about healthcare costs when there are border issues?"
TEST: Does this new topic actually address the original question?
8. TU QUOQUE (You Too / Whataboutism)
"You say smoking is bad, but you used to smoke."
TEST: Does the other person's behavior change the facts?
9. GENETIC FALLACY
Judging an idea based on its origin rather than its merit.
"That policy was proposed by [group I dislike], so it's bad."
TEST: Evaluate the idea on its own merits, regardless of source.
10. APPEAL TO IGNORANCE (Argumentum ad Ignorantiam)
"No one has proven it false, so it must be true."
TEST: Absence of evidence is not evidence of absence
(but it's also not evidence of presence).
Fallacies of Presumption
11. BEGGING THE QUESTION (Circular Reasoning)
The conclusion is assumed in the premise.
"God exists because the Bible says so, and the Bible is true
because it's the word of God."
TEST: Is the conclusion being used to support itself?
12. FALSE DICHOTOMY (Either/Or)
Presenting only two options when more exist.
"You're either with us or against us."
TEST: Are there really only two options? What's the middle ground?
13. HASTY GENERALIZATION
Drawing broad conclusions from insufficient evidence.
"I met two rude people from that city. Everyone there is rude."
TEST: Is the sample size and representativeness sufficient?
14. SLIPPERY SLOPE
Claiming one event will inevitably lead to extreme consequences
without evidence for each step in the chain.
"If we allow X, next it'll be Y, then Z, then total chaos."
TEST: Is each step in the chain actually inevitable?
15. LOADED QUESTION
A question that presupposes something unproven.
"When did you stop cheating on your taxes?"
TEST: Does the question contain an unproven assumption?
16. FALSE CAUSE (Post Hoc Ergo Propter Hoc)
Assuming that because B followed A, A caused B.
"I wore my lucky socks and we won the game."
TEST: Is there a demonstrated causal mechanism? Could it be coincidence?
17. EQUIVOCATION
Using a word with multiple meanings to confuse the argument.
"The sign said 'fine for parking here,' so I parked because it's fine."
TEST: Is the same word being used with different definitions?
18. COMPOSITION
Assuming what's true of parts is true of the whole.
"Every player on this team is excellent, so the team is excellent."
TEST: Does the property of parts necessarily transfer to the whole?
19. DIVISION
Assuming what's true of the whole is true of each part.
"This is the best university, so every professor here must be the best."
TEST: Does the property of the whole necessarily apply to each part?
Fallacies of Ambiguity and Weak Induction
20. STRAW MAN
Misrepresenting an opponent's argument to make it easier to attack.
Them: "We should have some regulations on X."
You: "They want to ban X entirely!"
TEST: Would the original person agree with your characterization?
21. NO TRUE SCOTSMAN
Redefining criteria to exclude counterexamples.
"No good manager would do that." "Bob did it." "Well, Bob
isn't a REAL manager."
TEST: Is the definition being changed to avoid counterevidence?
22. MOVING THE GOALPOSTS
Changing the criteria for proof after the original criteria are met.
"Show me a study." [Study shown.] "That study isn't large enough."
[Larger study shown.] "Well, it wasn't long enough..."
TEST: Were the criteria defined before the evidence was presented?
23. CHERRY PICKING (Suppressed Evidence)
Selecting only evidence that supports your position while
ignoring contradictory evidence.
"Studies show chocolate is healthy!" (ignoring the 10 studies
that show it isn't)
TEST: What does the FULL body of evidence say, not just selected pieces?
24. TEXAS SHARPSHOOTER
Finding a pattern after the fact and claiming it was predicted.
Drawing the bullseye around the bullet holes.
TEST: Was the prediction made before or after the data was collected?
25. GAMBLER'S FALLACY
Believing that past random events affect future probabilities.
"The coin landed heads 5 times, so tails is 'due.'"
TEST: Are the events actually independent?
26. APPEAL TO CONSEQUENCES
Arguing something is true/false based on its consequences.
"Climate change can't be real because the economic impact would
be devastating."
TEST: Does the desirability of a consequence affect its truth?
27. MIDDLE GROUND (Argument to Moderation)
Assuming the truth is always between two extremes.
"One says 2+2=4, another says 2+2=6, so maybe it's 5."
TEST: Is the middle position supported by evidence,
or just assumed to be correct because it's in the middle?
28. BANDWAGON (Argumentum ad Numerum)
"Everyone is switching to X, so you should too."
TEST: Is popularity evidence of quality?
29. ANECDOTAL EVIDENCE
Using personal stories instead of systematic evidence.
"My grandfather smoked and lived to 95, so smoking isn't that bad."
TEST: Does one story supersede systematic data from thousands of cases?
30. NIRVANA FALLACY
Comparing a real option to an unrealistic, perfect alternative.
"This plan isn't perfect, so we shouldn't do it."
TEST: Is the comparison to a realistic alternative or an impossible ideal?
31. SUNK COST FALLACY
Continuing because of past investment rather than future value.
"We've spent $2M on this project. We can't stop now."
TEST: If you hadn't already invested, would you start this today?
32. APPEAL TO NOVELTY
"This approach is new, so it must be better."
TEST: Is there evidence the new approach is actually better?
Evidence Evaluation Framework
Source Credibility Assessment
THE CRAAP TEST:
C - CURRENCY: When was this published/updated?
Is the information current enough for your topic?
Are the sources cited also current?
R - RELEVANCE: Does this actually address your question?
Is it at the right level of depth?
Who is the intended audience?
A - AUTHORITY: Who created this? What are their credentials?
Is the publisher reputable?
Are credentials verifiable?
A - ACCURACY: Is the information supported by evidence?
Can claims be verified through other sources?
Has it been peer-reviewed or fact-checked?
Are there spelling/grammar errors (a quality signal)?
P - PURPOSE: Why does this information exist?
Is it to inform, persuade, sell, entertain?
Are there obvious biases or conflicts of interest?
Is the tone objective or emotional?
Evaluating Research Studies
STUDY EVALUATION CHECKLIST:
METHODOLOGY:
[ ] What type of study? (RCT, observational, case study, meta-analysis)
[ ] Sample size adequate?
[ ] Control group present?
[ ] Was it randomized? Double-blinded?
[ ] Duration sufficient for the effect being measured?
[ ] Could confounding variables explain the results?
RESULTS:
[ ] Statistical significance vs. practical significance
(p < 0.05 doesn't mean the effect is large or important)
[ ] Effect size reported? (How big is the actual difference?)
[ ] Confidence intervals provided?
[ ] Were all outcomes reported? (Or just the favorable ones?)
[ ] Was the study replicated by independent researchers?
INTERPRETATION:
[ ] Do the conclusions follow from the data?
[ ] Are limitations acknowledged?
[ ] Is correlation being presented as causation?
[ ] Does the study apply to your specific situation?
[ ] Who funded the research? (Potential conflicts of interest)
HIERARCHY OF EVIDENCE (strongest to weakest):
1. Systematic reviews and meta-analyses
2. Randomized controlled trials (RCTs)
3. Cohort studies (observational, prospective)
4. Case-control studies (observational, retrospective)
5. Case series and case reports
6. Expert opinion and editorials
7. Anecdotal evidence
Argument Mapping
Building an Argument Map
AN ARGUMENT MAP visualizes the structure of reasoning:
MAIN CLAIM
|
|-- Supporting Reason 1
| |-- Evidence A
| |-- Evidence B
| |
| |-- [OBJECTION: Counter-evidence C]
| |-- [REBUTTAL: Response to C]
|
|-- Supporting Reason 2
| |-- Evidence D
| |-- Evidence E
|
|-- [OBJECTION: Alternative Explanation]
|-- [REBUTTAL: Why alternative fails]
EXAMPLE:
CLAIM: "Remote work increases productivity"
|
|-- Reason: Fewer interruptions
| |-- Stanford study: 13% productivity increase
| |-- Survey data: 67% report fewer distractions
| |
| |-- [OBJ: Some roles need in-person collaboration]
| |-- [REB: Hybrid models address this]
|
|-- Reason: Eliminated commute time
| |-- Average 40 min saved per day
| |-- 60% report using saved time for work
|
|-- [OBJ: Collaboration and innovation suffer]
|-- [REB: Meta-analysis shows mixed results;
depends on work type and tools used]
BENEFIT: Argument mapping forces you to see the STRUCTURE
of reasoning, making weak points and gaps visible.
Scientific Thinking
THE SCIENTIFIC METHOD FOR EVERYDAY THINKING:
1. OBSERVE: Notice something interesting or puzzling
2. QUESTION: Formulate a specific, testable question
3. HYPOTHESIZE: Propose an explanation
4. PREDICT: If the hypothesis is true, what should we observe?
5. TEST: Design an experiment or gather data
6. ANALYZE: What does the evidence show?
7. CONCLUDE: Does the evidence support or refute the hypothesis?
8. COMMUNICATE: Share findings and methods for scrutiny
KEY SCIENTIFIC THINKING PRINCIPLES:
FALSIFIABILITY:
A claim must be testable and potentially provable wrong.
"There's an invisible dragon in my garage" -- unfalsifiable
"This drug reduces blood pressure by 10%" -- falsifiable
PARSIMONY (OCCAM'S RAZOR):
Among competing hypotheses that explain the data equally well,
prefer the one with the fewest assumptions.
REPRODUCIBILITY:
Results should be obtainable by others using the same methods.
One study is a data point. Replication is where confidence builds.
NULL HYPOTHESIS:
Start from the assumption that there is no effect.
The burden of proof is on the claim, not on the skeptic.
BAYESIAN UPDATING:
Adjust your confidence based on new evidence.
Strong evidence should change your mind significantly.
Weak evidence should change it slightly.
No evidence should not change it at all.
Media Literacy
EVALUATING NEWS AND MEDIA:
1. CHECK THE SOURCE
- Is it a known, reputable outlet?
- Does the outlet have an editorial/fact-checking process?
- What is the outlet's known bias? (all outlets have some)
2. CHECK THE AUTHOR
- Do they have expertise or a track record in this area?
- Can you find their other work?
- Are they reporting or editorializing?
3. CHECK THE EVIDENCE
- Are specific sources cited?
- Can you verify the claims independently?
- Is the original source accessible?
4. CHECK THE FRAMING
- What's the headline vs. what the article actually says?
- Are there qualifying words (might, could, suggests) that
the headline drops?
- Is important context missing?
5. CHECK YOUR REACTION
- Did this make you feel a strong emotion (outrage, fear, triumph)?
- Strong emotional reactions are a signal to slow down and verify
- Are you sharing this because it's true or because it confirms
what you already believe?
MISINFORMATION PATTERNS:
- Out-of-context images/quotes
- Misleading statistics (true numbers, wrong framing)
- False attribution (real quote, wrong person)
- Satire mistaken for news
- Old stories presented as current
- Manipulated images/video
- Astroturfing (fake grassroots movements)
Biases That Impair Critical Thinking
CONFIRMATION BIAS:
Seeking and favoring information that confirms existing beliefs.
Antidote: Actively seek disconfirming evidence. Ask: "What would
change my mind?"
ANCHORING BIAS:
Over-relying on the first piece of information encountered.
Antidote: Seek multiple reference points before forming judgment.
AVAILABILITY HEURISTIC:
Judging probability by how easily examples come to mind.
Antidote: Check base rates. Dramatic events feel common but may be rare.
DUNNING-KRUGER EFFECT:
Those with low ability overestimate their competence; experts
tend to underestimate theirs.
Antidote: Seek feedback from those more knowledgeable. Calibrate.
MOTIVATED REASONING:
Using reasoning to justify a conclusion you've already reached.
Antidote: Ask "Am I reasoning toward truth or toward a preferred conclusion?"
GROUPTHINK:
Conforming to group consensus to maintain harmony.
Antidote: Assign a devil's advocate. Use anonymous input.
Seek outside perspectives.
SURVIVORSHIP BIAS:
Drawing conclusions only from successes, ignoring failures.
"Every successful CEO dropped out of college!"
Antidote: Ask about the base rate. How many college dropouts
DIDN'T become successful CEOs?
Practice Exercises
Exercise 1: Fallacy of the Day
Each day for a month, identify one logical fallacy in news, social media, or conversation. Write it down with the specific fallacy type and why it fails.
Exercise 2: Steel Man Challenge
Take a position you strongly disagree with. Construct the strongest possible argument FOR that position. This builds intellectual empathy and sharpens your ability to see weaknesses in your own arguments.
Exercise 3: Source Evaluation
Take a viral social media claim. Apply the CRAAP test. Trace it to its original source. Document what you find. How does the original compare to how it was shared?
Exercise 4: Argument Map
Take an editorial or opinion piece. Map its argument structure completely (claims, reasons, evidence, objections, rebuttals). Identify where the argument is strong and where it's weak.
Exercise 5: Pre-Mortem
Before making your next important decision, assume it failed. Work backwards to identify all the ways your reasoning could have been flawed. Which biases might have affected your analysis?
Exercise 6: Belief Audit
List 5 strong beliefs you hold. For each, write: What evidence would change my mind? When did I last update this belief? What is the strongest argument against it?
Output Format
Deliver the response as a structured document with clear headings and actionable content. Use tables for comparisons, numbered lists for sequential steps, and bullet points for options. Include specific examples where applicable.
[Critical Thinker deliverable]
1. Context and objectives
2. Analysis or framework
3. Specific recommendations with rationale
4. Action items with timeline
Example
Input: "Help me with critical thinker for a mid-size project."
Output: A complete critical thinker framework tailored to the specific context, with actionable steps, relevant considerations, and measurable outcomes.
Edge Cases
- Incomplete information: Ask clarifying questions before proceeding rather than making assumptions
- Conflicting requirements: Identify trade-offs explicitly and present options with pros and cons
- Scale mismatch: Adapt recommendations to match the user's context (individual vs. team vs. organization)
- Domain crossover: When the request overlaps with other skill domains, address what falls within scope and reference specialized skills for the rest