Start with the honest finding, because it changes the workflow. UniBanQ has AI, but not the AI this article's title implies. Its AI tooling is built for content transformation — turning documents into summaries, flashcards and podcasts — and for OSCE-style voice simulation with spoken feedback. It does not ship a conversational Socratic tutor you can interrogate about why option C beats option D on a missed multiple-choice question. If you came to UniBanQ expecting a per-item MCQ chatbot, that specific tool is not what it offers. This workflow shows you how to run a rigorous prompt-and-verification loop for every missed MCCQE Part I question anyway, using UniBanQ's written explanations and references as the material to verify — and where iatroX's Socratic Tutor fills the gap UniBanQ leaves.
What UniBanQ offers for the MCCQE Part I right now
Vendor-reported, last checked 19 July 2026. Confirm on unibanq.ai.
| Item | What UniBanQ states (vendor-reported, 19 July 2026) |
|---|---|
| MCCQE Part I question volume | The homepage cites "+3000 questions for MCCQE1"; the features page cites "+5000 questions adapted to your level" across all levels. Treat the MCCQE-specific figure as roughly 3,000 and verify it. |
| AI features | AI clinical simulators (voice-to-voice history taking, OSCE practice with spoken feedback) and a "Study Desk" that turns resources into summaries, flashcards, questions and podcasts. There is no dedicated Socratic tutor for explaining individual MCQs. |
| Access and pricing | Expert Learner at about $34.99 CAD per month, roughly $28.99 per month billed annually ($339.99 per year), or a lifetime plan around $799.99. Confirm current pricing at checkout. |
| Components supported | The written MCQ paper, with explanations tied to "the latest local guidelines" plus PubMed links and curated videos. The OSCE simulator addresses the separate clinical exam, not the MCCQE Part I MCQ paper. |
The consequence of the honesty flag is practical. Your "tutor" for missed MCQs is not a UniBanQ chatbot; it is (a) UniBanQ's own written explanation and its cited local guidelines and PubMed links, and (b) an external Socratic tutor such as iatroX's when you want to be questioned rather than told. The prompts below are written to be used against those, not against a UniBanQ feature that does not exist.
The exam you are actually training for
Since the 2025 change, the MCCQE Part I is multiple-choice only; the clinical decision-making component was removed. It is 230 MCQs in two sections of 115, each timed at two hours and forty minutes. Confirm the live count and timing on mcc.ca. MCC objectives sit under the CanMEDS roles across Dimensions of Care (health promotion and illness prevention, acute, chronic, psychosocial) and Physician Activities (assessment and diagnosis, management, communication, professional behaviours). The MCC's own practice products are the calibration standard; UniBanQ's explanations are third-party interpretation whose currency and Canadian basis you should verify per item.
A prompt taxonomy for verifying a missed question
Because you are prompting UniBanQ's written explanation (and, where you want questioning, an external tutor), the prompts are aimed at interrogating a rationale rather than generating one. Organise them into six kinds: mechanism (why the correct option is correct), discriminating features (which stem detail decides it), option elimination (why each distractor fails), guideline verification (which Canadian source and date supports the management step), counterfactuals (what change would flip the answer), and retrieval testing (a fresh unseen variant you answer before checking).
Twenty-five reusable prompts, grouped by the six commonest error types
Keep these generic and never paste proprietary UniBanQ question text into any external AI tool.
Knowledge gap (mechanism):
- "In three sentences, explain the mechanism linking [finding] to [diagnosis]."
- "What single principle does this item test?"
- "Where does UniBanQ's explanation state its source, and is it Canadian?"
- "Give two other presentations governed by this same mechanism."
Misread stem (discriminating features): 5. "Which exact phrase in the stem determines the answer?" 6. "What qualifier — 'most likely', 'initial', 'next' — did I need to weight?" 7. "If I had misread [detail], which distractor would I have chosen?" 8. "Restate the stem using only the details that change management."
Premature closure (option elimination): 9. "Give the specific reason each of the five options is right or wrong." 10. "What is the second-best answer and why is it second?" 11. "Which distractor is the trap for early anchoring?" 12. "Justify eliminating my first instinct before I confirm."
Guideline/jurisdiction error: 13. "Name the current Canadian guideline for this and its year." 14. "Does UniBanQ's cited guideline match current Canadian practice?" 15. "What threshold or screening interval does Canadian practice use here?" 16. "Is this reportable, or is there a Canada-specific consent nuance?"
Calculation error: 17. "Show the calculation step by step with units." 18. "Where is the likeliest arithmetic slip on this item?" 19. "Give me a structurally identical calculation to redo now."
Time-pressure/retrieval: 20. "Compress the reasoning into one recallable sentence." 21. "Give a 20-second heuristic for this item type." 22. "Pose an unseen variant and withhold feedback until I answer." 23. "Re-test me on this principle in three days." 24. "State the single highest-yield fact to retain." 25. "State the rule, then ask me to apply it to a scenario you invent."
The anti-answer-leak rule
Commit before you consult. Write your chosen option and a one-line rationale before you open UniBanQ's explanation or any external tutor. With a written explanation the leak risk is smaller than with a hint chatbot, but the discipline still matters: if you read the rationale while undecided, you convert a genuine test into passive recognition. Commit first; verify second; and if you cannot commit, log the item as a real knowledge gap rather than a near-miss.
A reusable jurisdiction-check prompt
UniBanQ says its explanations rest on "the latest local guidelines," which is encouraging but must be checked. Run this on every clinically specific item, against the written explanation and, if needed, an external tutor: "Does this rationale reflect current Canadian practice for the MCCQE Part I? Name the Canadian guideline it rests on and the date of that guidance; if the basis is non-Canadian or undated, say so." If neither the explanation nor the tutor can name a Canadian source and date, treat the answer as unverified and check a primary Canadian source before trusting it.
Create a misconception record
For each genuine correction, log four lines: the incorrect rule you held, the corrected rule, a transfer question testing the corrected rule on new terrain, and a review date. UniBanQ's Study Desk can help here in the one place its content-transformation AI is genuinely useful — turning your accumulating misconception records into flashcards or a spaced set — but the four-line discipline is yours, not the tool's.
Weekly verification sample
Any AI-generated or AI-curated explanation can be wrong or out of date, and "based on local guidelines" is a claim to audit, not to accept. Once a week, take a small random sample — around ten of UniBanQ's substantive claims from that week — and independently check them against MCC objectives, current Canadian guidelines and primary literature, logging every discrepancy. If the discrepancy rate rises, rely less on the explanations for unsupervised facts. This is the same audit iatroX describes for auditing an AI exam tutor and for calibrating automated feedback — and it applies to the spoken feedback from UniBanQ's OSCE simulator too.
A worked seven-day plan for an international graduate
Fatima, an IMG five weeks out, uses UniBanQ for one job — a large MCQ pool with guideline-linked written explanations — and iatroX for unseen transfer measurement and the Socratic questioning UniBanQ does not provide.
- Day 1: 40 mixed, timed UniBanQ items; commit-before-consult on every miss; code the error type.
- Day 2: Verify each miss against UniBanQ's explanation and its cited guideline; jurisdiction-check the clinically specific ones; write misconception records.
- Day 3: 40 items on management and communication; a transfer question for each principle, not a repeat.
- Day 4: A fresh, timed, mixed MCCQE Part I block in iatroX — genuinely unseen — to measure transfer.
- Day 5: Open a missed item in the iatroX Socratic Tutor and run the verification prompt before viewing the explanation — the questioning step UniBanQ cannot give you.
- Day 6: Weekly verification sample of UniBanQ claims against primary sources; one official MCC practice component for calibration.
- Day 7: Convert the week's misconception records into a spaced set (UniBanQ's Study Desk is fine for this); light review only.
UniBanQ supplies volume and guideline-linked explanations; iatroX supplies the unseen ruler and the Socratic questioning; the official material calibrates.
Decision checklist: continue, supplement, switch or stop
- Continue if UniBanQ's explanations are verifying cleanly, your misconception log stays fixed, and unseen iatroX blocks are rising.
- Supplement with the Socratic questioning UniBanQ lacks (iatroX or another genuine tutor) if you find that written explanations alone are not exposing your reasoning errors — some candidates need to be questioned, not told. The two-Q-bank rule keeps the second source from duplicating items.
- Switch or de-emphasise if your weekly verification sample shows a rising discrepancy rate or repeated non-Canadian defaults.
- Stop adding bank time when your exit criteria — coverage floors, stable unseen accuracy, pacing, retention, official-material agreement — are all met.
Decide on measured signals, not on the novelty of UniBanQ's podcast and simulator features.
Frequently asked questions
Is UniBanQ enough for MCCQE Part I on its own? For the written paper, UniBanQ offers a substantial MCQ pool (vendor-reported around 3,000) with guideline-linked explanations, so it can serve as a core bank — but "on its own" is limited by what it does not do: it has no Socratic MCQ tutor, so it tells rather than questions, and its explanations need Canadian-jurisdiction verification. Add official MCC practice material for calibration and an unseen bank plus genuine Socratic questioning for transfer and reasoning, and it becomes a stronger core than it is alone.
Which MCCQE Part I component does UniBanQ not reproduce well? UniBanQ reproduces the written MCQ format with explanations, but it does not reproduce interactive per-question tutoring — there is no conversational Socratic tutor for missed MCQs, only written rationales. Its clearly AI-forward features, the OSCE voice simulator and the content-transformation Study Desk, address the separate clinical exam and study-material creation respectively, not the reasoning-under-pressure of the MCQ paper, which you must rehearse with unseen, unassisted blocks.
How should I verify UniBanQ AI answers for MCCQE Part I? Verify in two layers. Per item, run the jurisdiction-check prompt and reject any clinically specific rationale that cannot be tied to a named, dated Canadian source, checking a primary source where UniBanQ's citation is vague. Weekly, independently audit a small random sample of UniBanQ's claims — and any spoken OSCE-simulator feedback — against MCC objectives and current Canadian guidelines, logging discrepancies and trusting the tool less as the discrepancy rate rises.
When should I stop using UniBanQ and move to mixed mocks? Move to mixed, timed, unassisted mocks once your blueprint cells are covered and your coded errors are mostly reasoning slips rather than knowledge gaps — not when you have exhausted the bank or the podcasts. Late in preparation the marginal gain comes from rehearsing the real format under a clock, so unseen mixed blocks and official practice material should displace filtered UniBanQ practice.
How should I combine UniBanQ with iatroX without duplicating practice? Give each a distinct job: UniBanQ for volume and guideline-linked written explanations; iatroX for fresh, timed, unseen blocks and for the Socratic questioning UniBanQ does not offer. Never re-use a UniBanQ item inside an iatroX session, and never run the iatroX Socratic Tutor on a question UniBanQ has already explained to you — the iatroX block's value is that it is unseen and unassisted. Duplicating items turns your measurement into a memory test.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026. UniBanQ figures (question counts, features, prices) are vendor-reported as displayed on unibanq.ai on that date and can change; note that UniBanQ's AI is oriented to content transformation and OSCE voice simulation rather than per-question MCQ tutoring, and its guideline-based explanations should be verified for Canadian currency per item. Confirm the exam format on mcc.ca. Disclosure: iatroX operates a competing MCCQE Part I question bank and a Socratic Tutor; both are confined here to jobs UniBanQ does not claim — clean, unseen, unassisted transfer measurement and Socratic questioning of missed items — and are not offered as a replacement for the official MCC materials or for UniBanQ's content volume. Corrections are welcome via the feedback route on iatrox.com. References: Medical Council of Canada MCCQE pages (mcc.ca); UniBanQ homepage and features pages; iatroX MCCQE Part I bank, the Socratic Tutor, the audit-an-AI-tutor pillar, and the percentage article.
Open the Socratic Tutor and test an unseen MCCQE Part I item in iatroX →