ProtocolTrack: AI-Generated Protocol Competency Quizzes
Upload your agency's protocol PDF once. Generate unlimited scenario-based competency quizzes from it on demand. Assign by user, certification level, or roster-wide. Track every provider's pass/fail status from one dashboard. Built for FTO programs, recertification cycles, state inspections, and any time you need defensible proof that your providers know your protocols.
Used by EMS agencies, fire department training divisions, nursing programs, and any clinical operation that needs to verify provider competency against published protocols.
What ProtocolTrack does
Upload once, quiz forever. Every feature is built around generating, assigning, and proving competency on your agency's protocols.
Upload your protocol PDF a single time. Replace it only when your medical director publishes a new version. All quizzes generate from this one source of truth.
Claude AI writes scenario-based questions grounded exclusively in your protocol text. No invented doses, no outside knowledge, no hallucinated guidelines.
Easy recall, application, full patient scenarios, or mixed. Set the percentage of scenario questions per quiz. Choose EMT-level or Paramedic-level content depth.
Assign quizzes to your full roster, by certification level (EMT, AEMT, Paramedic), or to specific individuals. Edit assignments without regenerating questions.
See exactly where every provider stands on every required quiz. Filter by status, sort by certification, export to CSV for your training records.
Per-quiz gradebook with attempts, best score, average score, and last activity. Best score is retained on retakes so providers can study and improve, right up until you release results, and you can drill into any provider's completed attempt question by question.
Randomize answer order per attempt, randomize question order, and set time limits with auto-submit. Each provider sees a unique version of every quiz.
Every quiz tied to the exact protocol version that generated it. CSV exports defend your training program for state inspections and accreditation reviews.
Per-quiz item difficulty, distractor analysis, and score distribution, plus point-biserial discrimination and reliability (Cronbach's alpha) once enough providers complete. Proof your test instrument is sound, not just that providers passed.
Results stay hidden until you release them. Providers retake to improve until you do, then the quiz locks and unlocks each provider's answers, correct responses, and protocol rationale for review.
Roster compliance, every provider, every required quiz
Filter by Not Started, Partial, Compliant, or None Required. See every provider's score across every assigned quiz. Export to CSV for state inspections, accreditation visits, or your medical director's quarterly review. No more chasing spreadsheets.
How It Works
From PDF upload to assigned competency quizzes in one short onboarding.
Drop your agency's current protocol document into ProtocolTrack. The system indexes it once and you are done.
Pick a topic focus, certification level, difficulty, and number of questions. Claude AI writes the quiz in seconds, grounded in your protocol text.
Assign to roster, role, or individual. Watch pass/fail roll in on the compliance dashboard. Export to CSV for inspections and recerts.
Real questions, written from your protocols
Every question references your specific protocol numbers. Mix of recall, application, and full patient scenarios at EMT, AEMT, or Paramedic level. The AI never invents doses or guidelines, every answer traces back to your uploaded protocol PDF.
Psychometric proof your test itself is sound
Every quiz comes with item analysis: difficulty per question, distractor analysis that surfaces which wrong answers are pulling responses (and the misconceptions behind them), score distribution, and pass rate. Once enough providers complete it, ProtocolTrack adds real psychometrics, item discrimination (point-biserial) and reliability (Cronbach's alpha and KR-20), so you can show each question separates strong performers from weak ones. The advanced stats appear only when there is enough data to be statistically valid. Until then the tool tells you how many more completions it needs instead of showing a number you cannot defend. That is psychometric evidence the instrument itself is sound, not just a roster of who passed.
You decide when results go live
Scores stay hidden by default. Providers see only pass or fail and their score, and they can retake to study and improve. When you are ready, you release results for the quiz. Releasing closes the quiz to further attempts and flips it into a teaching tool: every provider unlocks their own answers, the correct responses, and the rationale tied to your protocol. One clean workflow, retake to improve right up until release, then the quiz locks and becomes a review of what your protocols actually say. Hide results again anytime to reopen attempts on your terms.
Every wrong answer becomes a lesson in your protocol
Once you release results, each provider walks their attempt question by question: what they picked, whether it was right or wrong, the correct answer, and the rationale that traces straight back to your protocol text. It turns a graded quiz into targeted protocol education with no extra work from you. Admins get the same per-question view into any provider's completed attempt, so a quick gradebook drill-down shows exactly where a provider went sideways and why.
Who It's For
Four roles, one shared dashboard, all anchored in your protocol document.
Generate, assign, and track competency quizzes from one dashboard.
Verify new hires know your protocols before they ride solo.
Defensible proof of provider competency for every credentialed provider.
Roster compliance reporting for state inspections and accreditation.
Built for EMS
Engineered with the same attention to detail and privacy posture as the rest of the platform.
Question generation powered by Claude with retrieval-augmented generation against your protocol PDF.
Voyage voyage-3-large embeddings and HNSW vector search retrieve only the relevant protocol chunks per question.
Every quiz is tied to the exact protocol version that generated it. Replace the PDF without losing existing quizzes.
Anthropic Batch API and prompt caching keep per-quiz cost low. Built to scale across agencies.
Provider data lives in your agency's private environment. Row-level security on every table.
ProtocolTrack only handles your protocol document and provider roster. No patient data ever ingested.
ProtocolTrack Pricing
AI-generated protocol competency quizzes for fire departments and EMS agencies. Built for FTO, recerts, and state inspections.
- Unlimited app users
- Unlimited AI-generated quizzes from your protocol PDF
- Targeted assignments by user, role, or certification level
- Roster compliance dashboard with CSV export
- Per-quiz gradebook with attempt tracking
- Anti-cheating controls (randomized answers and questions, time limits)
- Versioned protocol document with audit trail
- Dedicated support
* Pricing above is per agency. Custom enterprise pricing for 25+ locations and multi-agency systems, see below.
Running 25 or more locations?
Departments with 25+ stations, apparatus, or storage points, plus county, regional, and multi-agency systems, are priced as a conversation, not a checkout. We'll scope it to fit.
Bundle this module with others for a discount. See all bundle options.
Pairs well with
Modules that ship most often alongside ProtocolTrack.