Now in institutional beta at Brigham Young University

Source code plagiarism detection built for how programming is actually taught

Honor Checker fingerprints and compares student code to surface real matches — across sections and prior terms, resistant to renaming and reformatting — and gives instructors side-by-side evidence they can actually stand behind.

Match · Section 2 vs. Section 4 92% match
student_A.java
int[] merge(int[] a, int[] b) {
int[] out = new int[a.length + b.length];
int i = 0, j = 0, k = 0;
while (i < a.length && j < b.length) {
if (a[i] <= b[j])
out[k++] = a[i++];
else
out[k++] = b[j++];
}
while (i < a.length) out[k++] = a[i++];
return out;
}
student_B.java
static int[] combine(int[] x, int[] y) {
int[] res = new int[x.length + y.length];
int p = 0, q = 0, r = 0;
while (p < x.length && q < y.length) {
if (x[p] <= y[q])
res[r++] = x[p++];
else
res[r++] = y[q++];
}
// fill remaining
while (q < y.length) res[r++] = y[q++];
}

Illustrative comparison — renamed variables and reordered lines don’t hide the shared logic.

Built on published research. Detection uses the Winnowing fingerprinting algorithm (Schleimer, Wilkerson & Aiken, SIGMOD 2003).

Broad language coverage. Deep, language-aware analysis for 14 languages, and fingerprinting for any other text-based source.

Now in institutional beta at Brigham Young University.

How it works

From upload to defensible evidence

  1. 1

    Bring in submissions

    Connect Honor Checker to an institutional course-management system — Instructure’s Canvas is supported, with others added based on demand — or upload single student submissions or submission batches by course section. Processing runs in the background with live progress.

  2. 2

    Normalize & fingerprint

    Code is normalized via AST transformation, then fingerprinted with the Winnowing algorithm — so cosmetic edits don’t hide copied logic.

  3. 3

    Review the evidence

    A ranked results table and side-by-side view highlight matching regions in both students’ code, with match percentages and filters.

Everything an integrity case actually needs

Real matches, fewer false positives, and evidence that holds up in a review.

Evasion-resistant detection

Renamed variables, reformatted whitespace and reordered imports don’t defeat it — code is normalized before fingerprinting.

Boilerplate exclusion

Instructor-provided starter code is fingerprinted and excluded, and a “de facto boilerplate” threshold suppresses code shared across many submissions — dramatically fewer false positives.

Cross-section & prior-term detection

Comparisons span every section and academic period of a course, catching submissions recycled from earlier terms or other sections.

Side-by-side evidence

Matched regions highlighted in both students’ code, with match percentages, filtering and sorting — evidence instructors can stand behind.

Built for institutions

Organizations with hierarchy, courses, academic periods, sections and projects; role-scoped admins; invitation-based onboarding.

Secure by design

AWS serverless architecture, per-tenant data isolation, Cognito authentication with required MFA, encrypted in transit.

Learn more

For administrators & deans

Uphold academic-integrity policy with detection built for the realities of programming courses: fewer false positives from shared scaffolding, matching that spans sections and prior terms, and an organizational model with role-scoped access that fits how your institution is structured.

For faculty & instructors

Upload a batch of submissions and get back a ranked, filterable list of the pairs worth a closer look — with matched regions highlighted side by side. Spend your time reviewing real evidence instead of eyeballing hundreds of files.

Roadmap · not yet shipped

What’s coming: AI assistance

The academic-integrity market was burned by unreliable “AI detectors.” Honor Checker’s AI story is built on the opposite premise — AI helps instructors interpret the evidence; it never renders a verdict.

Deterministic evidence first, AI assistance second.

Matches come from a published, reproducible fingerprinting algorithm — not an AI guess. When AI features ship, they interpret, triage, and draft; a human always decides. AI never accuses a student.

Get notified when Honor Checker is available

Honor Checker is in institutional beta. Leave your details and we’ll email you the moment it’s available for your institution.