In development for Windows and Mac
Skopos is a free desktop study app for the MCAT. It builds each day around your exam date, teaches from open textbooks, tests you with exam-format questions and brings every flashcard back shortly before you would forget it.
Features
Reading, questions, flashcards and practice exams sit in one plan. Press Start and the next thing is ready.
Plan
Skopos builds a day-by-day schedule from your exam date and the hours you have, with catch-up and rest days. Miss a day and it offers three ways back: catch up slowly, study longer each day, or move the exam date.
The plan is not stored. It is recalculated from your real progress every time, so it does not go stale.
Questions
Passage-based questions in the AAMC format, with figures, equations and a clock. The toolbar has highlight, strikethrough, flag for review and the periodic table, as in the official test layout.
Explanations
Every wrong answer is explained step by step: why the option you chose does not hold, and why the correct one works. In a blind review, these explanations scored 4.87 out of 5.
Flashcards
Rate how well you knew a card and Skopos schedules the next review with FSRS, an open spaced-repetition algorithm. Each button shows when the card will come back.
Anki decks import and export. A deck of over 6,000 notes imports in under 30 seconds.
Sources
Science content comes from OpenStax textbooks. Each card shows the section it came from, and every textbook figure carries its source and licence. The built-in reader opens the full section.
Progress
Progress shows how much of each topic you are likely to remember today, and which topics need a review. One button turns that list into a short session.
Practice exams
Sit a full-length exam or a single section under exam conditions, with timed breaks and no going back to a closed section. Plan one for a set day and it joins your schedule.
Try a card
Read the question, recall the answer, then flip the card. Select a card or press Enter or Space to flip it.
These eight cards come from the starter set that ships with the app. Each one passed the automated card checks and a separate review. B/B is biology and biochemistry, C/P is chemistry and physics, P/S is psychology and sociology.
Try a passage
Two passage sets written by the Skopos question generator, one in chemistry and physics and one in critical analysis and reasoning. Every question passed the generator's checks, and five of the eight also passed a blind read by two reviewers. Choose an answer to see the explanation the app gives.
Hydrostatic pressure reflects the force per unit area exerted by the weight of a static fluid column. In medical applications such as open-tube manometers and intravenous (IV) infusion lines, the hydrostatic gauge pressure generated by a fluid column of height h and density ρ under gravitational acceleration g is expressed by Equation 1. Because the free surface of the fluid in these open systems is exposed to the atmosphere, atmospheric pressure acts on that surface, so the absolute pressure at depth h equals the gauge pressure plus atmospheric pressure, as defined by Equation 2.
In Experiment 1, researchers measured the hydrostatic gauge pressure (Pg) at various depths h below the open fluid surface within vertical cylindrical columns open to the atmosphere (Patm = 1.01 × 105 Pa, g = 9.8 m/s2). Two test fluids were evaluated: normal saline (ρ = 1.00 × 103 kg/m3) and an iodinated radiographic contrast solution (ρ = 1.36 × 103 kg/m3). As Table 1 shows, gauge pressure was recorded at depths of 0.25 m, 0.50 m, and 1.00 m.
In Experiment 2, researchers connected the saline reservoir to a compliant vessel simulator maintained at an internal venous gauge pressure of 2400 Pa (equivalent to approximately 18 mm Hg). The vertical height of the reservoir's fluid surface above the vessel entry point was adjusted incrementally to determine the minimum height required to initiate flow from the reservoir into the vessel. Fluid entry into the vessel commenced only when this vertical height equaled or exceeded 0.245 m.
| Fluid | Density (kg/m3) | Depth h (m) | Gauge Pressure Pg (kPa) |
|---|---|---|---|
| Saline | 1.00 × 103 | 0.25 | 2.45 |
| Saline | 1.00 × 103 | 0.50 | 4.90 |
| Saline | 1.00 × 103 | 1.00 | 9.80 |
| Contrast agent | 1.36 × 103 | 0.25 | 3.33 |
| Contrast agent | 1.36 × 103 | 0.50 | 6.66 |
| Contrast agent | 1.36 × 103 | 1.00 | 13.33 |
Pg = ρ g h
Equation 1
Pabs = Pg + Patm
Equation 2
Adapted from College Physics, OpenStax, CC BY 4.0.
Varying fluid density in uniform cylindrical columns tests the effect of density rather than vessel geometry, replicating the comparison between saline and contrast agent already conducted in Experiment 1.
Sealing the columns alters the surface boundary condition and ambient pressure transmission, failing to evaluate whether cross-sectional geometry affects hydrostatic gauge pressure.
Allowing fluid depths to vary when changing column diameter introduces depth as an uncontrolled confounding variable, preventing direct isolation of container geometry.
Hydrostatic gauge pressure depends on fluid density, gravitational acceleration, and depth below the surface according to Equation 1 (Pg = ρgh), independent of vessel shape or cross-sectional area. To test whether column geometry influences the pressure measurements, the experimental design must vary the container geometry while controlling all other variables by holding depth, density, and ambient exposure constant.
This value represents the reduction in required height relative to saline (Δh = 0.245 m - 0.18 m = 0.065 m ≈ 0.07 m), confusing the difference between the two configurations with the absolute height required for the contrast agent.
Equation 1 indicates that hydrostatic pressure depends linearly on fluid density (Pg = ρgh); scaling the column height inversely with the square of relative density rather than its first power produces 0.13 m.
This value arises from treating column height as directly proportional to fluid density (multiplying 0.245 m by 1.36) rather than recognizing the inverse relationship h = Pg / (ρg).
To initiate fluid flow into the vessel simulator, the hydrostatic gauge pressure generated by the fluid column must reach the opposing venous gauge pressure of 2400 Pa. Rearranging Equation 1 (Pg = ρgh) for column height gives h = Pg / (ρg). Substituting the density of the iodinated contrast agent from Table 1 (1.36 × 103 kg/m3) and gravitational acceleration (9.8 m/s2) yields h = 2400 Pa / ((1.36 × 103 kg/m3)(9.8 m/s2)) ≈ 0.18 m.
This choice mistakes the hydrostatic gauge pressure recorded in Table 1 (9.80 kPa = 9.80 × 103 Pa) for absolute pressure, omitting the atmospheric pressure term required by Equation 2.
This value results from subtracting the fluid column's gauge pressure from atmospheric pressure (1.01 × 105 Pa − 9.80 × 103 Pa), whereas absolute pressure requires adding the two pressures.
This choice considers only atmospheric pressure acting at the open surface (1.01 × 105 Pa), neglecting the additional hydrostatic pressure exerted by the 1.00 m column of saline.
Equation 2 defines absolute pressure as the sum of gauge pressure and atmospheric pressure (Pabs = Pg + Patm). In Experiment 1, atmospheric pressure is given as 1.01 × 105 Pa, and Table 1 reports the hydrostatic gauge pressure at a depth of 1.00 m in the saline column as 9.80 kPa (9.80 × 103 Pa). Adding the atmospheric pressure to this hydrostatic contribution gives 1.01 × 105 Pa + 0.098 × 105 Pa = 1.11 × 105 Pa.
This reflects the misconception that hydrostatic pressure follows a quadratic dependence on depth; however, Equation 1 indicates a first-order linear relationship with column depth.
This value arises from mistakenly computing the fractional increase in absolute pressure by including atmospheric pressure (from approximately 107.7 kPa to 114.3 kPa), whereas the question asks specifically for gauge pressure.
This relies on the belief that a constant fluid density prevents pressure from varying; however, hydrostatic gauge pressure scales directly with depth regardless of uniform density.
Equation 1 defines hydrostatic gauge pressure as Pg = ρgh. Because density ρ and gravitational acceleration g are constant, gauge pressure is directly proportional to depth h. Doubling the depth from 0.50 m to 1.00 m doubles the gauge pressure, as reflected in Table 1 where Pg increases from 6.66 kPa to 13.33 kPa.
Most people who care about old buildings assume that decay is the enemy and that the craftsman who replaces a rotten lintel does the same kind of good as the physician who sets a bone. The assumption is generous, and it is mistaken. A building is not merely a shape that happens to be old; it is a shape that carries the evidence of its own making and of every season it has survived, and that evidence cannot be restored once it has been removed.
Consider what happens when a weathered cornice is cut away and a fresh stone is set in its place. The new block may match the old in profile and in colour. It will not match it in the one quality that mattered: it was not cut by the hands that cut the rest, in the year the rest was cut, to do the work the rest was doing. What the visitor now sees is a modern object wearing an old building's face, and the substitution is hardest to detect exactly where it is most complete.
Defenders of the practice answer that the alternative, watching a building come apart, betrays the very object they claim to love. This is the strongest objection, and it deserves an answer rather than a shrug. In part it is right: grime is not patina, a leaking roof ruins the plaster beneath it, and a masonry wall left to drink water will not be preserved by reverence. Maintenance is not surrender. A building kept in repair, re-pointed and drained and re-roofed, is still the building, because repair keeps what is there instead of substituting for it.
The difficulty begins where maintenance ends. At some point a wall is no longer a wall but a pile of stones agreeing to stand, and then the choice is not between the original and a copy. It is between a dignified end and a counterfeit that will be mistaken for the original for a century. The counterfeit is worse, because it destroys the possibility of honest mourning. A ruin tells the truth about time; a restored ruin tells a comfortable lie, and the lie is told in the building's own voice.
Consider the great houses that have been rebuilt wing by wing, each generation quietly replacing what the last had saved. Stand in the courtyard and you see a building older than any of its parts. The guide will give you its date, and the date will be false. Nothing was lost in a single blow; everything was lost by degrees, which is how most losses of this kind happen. Each repair was defensible on its own, and the sum of defensible repairs was a new building with an old reputation.
What the building deserved was attention rather than resurrection. The honest duty is modest and continuous: keep the weather out, keep the mortar sound, and when the structure finally fails, let it fail visibly, so that the record it carried is not confused with the copy that replaced it. To save an appearance is not to save a building. Restoration destroys more completely than neglect, because neglect leaves a true thing broken while restoration leaves a false thing whole.
Minor point not main: the cornice is the passage's evidence for the substitution, and the closing sentence states the comparative claim the example supports, so the example is not the argument.
Too extreme: the passage says maintenance is not surrender and that a building kept in repair is still the building, so its caution about restoration never reaches maintaining a building.
True but irrelevant: the passage does make that distinction, but it serves the concession to the defenders rather than stating the argument the passage as a whole advances.
Foundations of Comprehension: the passage opens by naming the common assumption that decay is the enemy and calling it mistaken, offers the replaced cornice as evidence that substituted work erases the record of making, concedes that grime is not patina, and closes by ranking the two harms directly: neglect leaves a true thing broken while restoration leaves a false thing whole. The thesis is that comparative verdict, and every paragraph works toward it.
Not in passage: the objection to the counterfeit is that it is trusted as a record it is not, and the passage never compares its cost with the cost of maintenance.
Too extreme: the passage describes substituted stones set among the rest, not a building stripped of every original part, so total replacement goes beyond the text.
Not in passage: the passage says the counterfeit will be mistaken for the original, which means those who rebuilt it present it as the original rather than as a forgery.
Foundations of Comprehension: the turn sets a dignified end against a counterfeit that will be mistaken for the original for a century, and the cornice paragraph explains what the substitution is: a new block matching the old in profile and colour but not cut by the hands that cut the rest. A counterfeit is therefore work presented as the original record, which is what the visitor trusts and what the author says cannot be recovered.
Minor point not main: the false date is the example's illustration of the loss, and the sentence after it states the point the example is there to make.
Not in passage: the example turns on how many repairs accumulate, not on the size of a house, and the passage nowhere compares large houses with small ones.
Contradicts passage: the example's verdict is that the accumulated repairs produced a new building with an old reputation, which is the loss the author regrets, not a practice being praised.
Reasoning Within the Text: the example follows the turn's claim that the counterfeit is worse because it destroys the possibility of honest mourning. Each generation's repair was defensible on its own, yet nothing was lost in a single blow and everything was lost by degrees, so the sum of defensible repairs is a new building with an old reputation. The example's function is to show cumulative loss rather than one act of destruction.
Contradicts passage: the passage locates the harm in substituted work, not in what observers perceive, so visitors' impressions cannot overturn a claim about the building's record.
Not in passage: nothing in the passage treats public interest as a licence for restoration, and its concession licenses maintenance rather than substitution.
True but irrelevant: the passage does say a rebuilt house is misdated, but misleading dating is a consequence it mentions and not the argument a survey of visitors would bear on.
Reasoning Beyond the Text: the author's verdict rests on what the building itself carries, not on what anyone feels in front of it. The first paragraph defines a building as a shape carrying the evidence of its own making and of every season it has survived, and the concession distinguishes maintenance from substitution on that ground. A survey of visitors' impressions reports felt authenticity, which the argument never makes its standard, so the verdict stands.
Research and evidence
Skopos uses learning methods with strong research support. Its content is tested blind against the materials students already use.
Reviews spread over time are remembered longer than the same reviews crammed together. Skopos schedules every card with FSRS, so each one returns at the point it is about to fade.
Cepeda, Pashler, Vul, Wixted and Rohrer (2006), Psychological Bulletin. Cepeda and colleagues (2008), Psychological Science.
Pulling an answer from memory strengthens it more than reading it again. New topics are taught in short parts, each followed by recall prompts, then practised with cards and questions.
Bjork and Bjork (2011), Making things hard on yourself, but in a good way, in Psychology and the Real World.
Memory consolidates between sessions, and long unbroken stretches add errors. Rest days and breaks are scheduled by clear rules, each with its stated reason and the research behind it.
Rasch and Born (2013), Physiological Reviews. Ackerman and Kanfer (2009), Journal of Experimental Psychology: Applied.
Every science question and card starts from one section of an OpenStax textbook. Anything that cannot be traced back to its section is discarded.
Biology and biochemistry questions mixed with AAMC and UWorld items, sources hidden, scored 1 to 5 by two independent raters.
96 cards across six topics, presented identically without labels, scored 1 to 5 by two independent raters.
| Questions | Skopos | UWorld | AAMC |
|---|---|---|---|
| Explanation | 4.87 | 4.10 | 3.08 |
| Answer key | 4.73 | 4.80 | 3.78 |
| Reasoning | 4.91 | 5.00 | 4.53 |
| Reads like the exam | 4.96 | 5.00 | 4.95 |
| Flashcards | Skopos | MileDown | AnKing-style |
|---|---|---|---|
| Correct | 4.60 | 3.92 | 3.56 |
| Depth | 3.02 | 2.13 | 2.71 |
| Pictures | 3.83 | 3.35 | 2.35 |
| Study value | 3.79 | 3.02 | 2.81 |
How Skopos differs
Most study apps today are cloud subscriptions tied to one AI service, and they keep your data on their servers. Skopos takes the opposite approach.
Study records sit close to health information, and that kind of data should not leave your computer without your say. Skopos does not use cloud storage. Data does not leave the computer unless you turn that on.
AI features are optional and use your own key. Skopos works with Anthropic, OpenAI, Google Gemini, Mistral, Groq and Ollama, which runs models on your own computer. A router picks the model for each task within the provider, quality setting and spending limit you chose. If no model meets every need, it tells you plainly and says what would fix it.
Built by
Bodi leads Skopos from Winnipeg. He defines the requirements and quality standards, designs the interface and tests every release. His biochemistry degree lets him verify the science content directly.
The code is written by a team of AI coding agents, Claude Code and OpenAI Codex, each with a defined role in planning, building, design or review. Bodi sets the rules and the review process they work to.
FAQ
Yes. Skopos is free, and it does not require an account.
Skopos is in development, with its first release planned for pre-med students at the University of Manitoba. Select Get launch updates to hear when it is ready.
Skopos is a desktop app for Windows and Mac. It works fully offline.
AI features are optional. When you use them, they run on your own key with the provider you choose, or on your own computer through Ollama.
Science content comes from OpenStax textbooks, and every science question and card traces back to the section it came from. Reading passages are written by the Skopos generator and checked by the same quality gates.
Yes. Skopos imports and exports Anki decks. A deck of over 6,000 notes imports in under 30 seconds.
On your computer, in an encrypted database. Skopos does not send it anywhere unless you turn that on.
No. MCAT is a registered trademark of the Association of American Medical Colleges. Skopos is not affiliated with or endorsed by the AAMC.