Paste any text. Get a structured analysis of its claims, premises, validity, and hidden assumptions — in seconds.
Free account · Upgrade from £10/mo
Think of it like a spell-checker, but for reasoning.
A spell-checker tells you every word is spelled correctly. It can't tell you whether what you wrote is worth reading. This tool tells you whether an argument holds together. It can't tell you whether it's right.
Every text runs through the same rigorous framework — adapted to the type of content, but always systematic. The method dates back to Aristotle. The tool adds speed and consistency.
Identifies what the text is ultimately claiming — whether stated outright or implied between the lines.
Maps the premises, both stated and hidden, and traces how they connect to support the conclusion.
Where it adds clarity, translates the argument into symbolic logic to reveal hidden dependencies and structural gaps.
The core question: does the conclusion actually follow from the premises? Evaluates deductive, inductive, and analogical reasoning.
Surfaces hidden assumptions and assesses whether the premises are likely to be true, uncertain, or require domain expertise to verify.
Checks for fallacies, ambiguity, missing qualifications, and counterarguments the author hasn't addressed.
A clear, honest verdict on the argument's logical quality — strengths, weaknesses, and how it could be strengthened.
Evaluates whether conclusions actually follow from premises — across deductive, inductive, analogical, and abductive reasoning.
Surfaces the unstated premises your argument relies on. Often, this is where the real disagreement lives.
Flags formal and informal fallacies with context — because not every pattern is a fallacy, and labels without explanation are useless.
Translates arguments into predicate notation when it reveals structure. Omits it when it wouldn't help. The formalization serves you, not the other way around.
The tool tells you what it can't assess. "Cannot verify — requires domain expertise" is more useful than false confidence.
Inline explanations of key concepts. Every report is also a lesson in how to read arguments more carefully.
The best outcome isn't that you use this tool forever. It's that using it teaches you to think this way naturally. This is a 2,400-year-old discipline that requires nothing more than careful reading, logical reasoning, and familiarity with common reasoning errors.
What the tool adds is speed and consistency. What you bring that the tool cannot is domain expertise, contextual understanding, and genuine comprehension of the subject matter.
The AI applies learned patterns from philosophy and logic. It functions like a well-trained philosophy graduate student — knowledgeable and systematic, capable of errors a formal theorem prover wouldn't make. We tell you this upfront.
Symbolic logic in reports is illustrative, not computationally verified. The value is in clarifying structure and surfacing issues, not in providing guarantees.
It cannot prove truth. It cannot replace domain expertise. It cannot resolve genuine substantive disagreements. It cannot guarantee its own correctness. These are genuine limitations, not fine print.
Does the conclusion follow?
An argument is valid when, if the premises were true, the conclusion would have to be true. Validity says nothing about whether the premises actually ARE true. A valid argument with false premises gives you a worthless conclusion, perfectly packaged.
Does it follow AND is it true?
Soundness is the gold standard: valid argument + all true premises = guaranteed true conclusion. But assessing soundness requires verifying premises, which often requires domain expertise the tool doesn't have. Treat soundness assessments as preliminary, not final.
The reasons given (and not given)
Premises are the starting points — the reasons offered for a conclusion. Some are stated. Others are hidden assumptions the author takes for granted. Hidden premises are often where the real action is.
Where reasoning goes wrong
A fallacy is an error in reasoning — structural or subtle. Not every instance of a pattern is fallacious. Context matters. And the absence of named fallacies doesn't mean the argument is good. There are ways to be wrong that don't fit neatly into categories.
Analyse shorter texts
Limited analyses per month
Full 7-section framework
Inline educational content on every report — learn the concepts as you go
Perfect for learning argument analysis
100 analyses per month
Full-length text support
Full 7-section framework
All text types — arguments, journalism, specs, marketing, satire
Priority processing
Higher volume analysis
Custom configurations
Team and enterprise options
API access available by discussion
Tailored to your needs
All plans include the same rigorous analysis framework. Free accounts include extra educational context to help you learn the discipline yourself.
Paste any text and get a structured analysis in seconds. We're putting the finishing touches on things — launching soon.
Launching soon · Free forever · Upgrade from £10/mo for full analysis