Use case · LMS and edtech

Plagiarism and AI detection API for LMS and edtech platforms

Check student submissions for plagiarism and AI writing from your own platform. Send the text from your backend, then show teachers each matching source and the sentences flagged as likely AI-written.

  • 2,000 free API words, no card
  • Free sandbox keys
  • REST + JSON
Integration pattern

Submit, poll, show the teacher

There is no LMS plugin or LTI app: you add checks in your platform’s backend with four REST calls.

  1. Student submits

    Your platform saves the submission as usual and extracts its text. The API takes plain text, not files.

  2. Your backend submits a scan

    A background job sends the text to POST /scans with the checks you want, and stores the scan ID with the submission.

  3. A worker polls for the result

    Poll GET /scans/{id} every few seconds until it is completed (usually 30–90 seconds), then save the result.

  4. The teacher reviews

    Show the plagiarism score with each source, and the AI score with the flagged sentences highlighted in the text.

Example

A queue worker that checks a submission

Shortened Node.js. The same pattern works in any language; open-source clients for seven languages are on the developer hub.

Submit, store the scan ID, poll until done
// Node.js 18+: run in a queue worker, never in the browser.
const BASE = "https://app.plagiarismcheckerplus.com/api/public/v1";
const headers = {
  Authorization: `Bearer ${process.env.PCP_API_KEY}`, // pcp_test_… while you build
  "Content-Type": "application/json",
};
const sleep = (ms) => new Promise((r) => setTimeout(r, ms));

async function call(url, init) {
  for (;;) {
    const res = await fetch(url, { headers, ...init });
    if (res.status === 429) {                       // rate limited: wait and retry
      await sleep(Number(res.headers.get("Retry-After") || 5) * 1000);
      continue;
    }
    const body = await res.json();
    if (!res.ok) throw Object.assign(new Error(body.error?.message), { code: body.error?.code });
    return body;
  }
}

export async function checkSubmission(submission) {
  const words = submission.text.trim().split(/\s+/).length;
  if (words < 50) return { skipped: "too short for AI detection" }; // 30 for plagiarism only

  const scan = await call(`${BASE}/scans`, {
    method: "POST",
    body: JSON.stringify({
      text: submission.text,
      tools: ["plagiarism", "ai_detect"],
      name: `submission-${submission.id}`,
    }),
  });
  await saveScanId(submission.id, scan.id); // your database

  let result;
  do {
    await sleep(5000);
    result = await call(`${BASE}/scans/${scan.id}`);
  } while (result.status === "pending" || result.status === "processing");

  return result; // results.plagiarism.sources[], results.aiDetection.sentences[]
}

Every result field is explained on the plagiarism checker API and AI detector API pages. Use offset and length on each flagged sentence to highlight it in the submission without re-matching text.

In practice

What to handle in an edtech integration

Skip short answers

Plagiarism needs 30 words and AI detection 50. Shorter text returns 400 text_too_short and costs nothing, but checking the count first saves a request.

Budget words per tool

Each tool uses the full word count: a 1,000-word essay checked for both uses 2,000 words. GET /account returns your remaining words and the longest text your plan accepts.

Spread out a class set

60 requests a minute per key on paid plans. Queue submissions, honour Retry-After on 429, and give each environment or tenant its own key (up to 10).

Build on test keys

A pcp_test_ key validates requests like a live one and returns sample results marked "sandbox": true, at no cost. Use it in development and CI.

Keep keys on the server

Call the API from your backend only. Never send a key to a browser, a mobile app or an LMS page script.

Handle failures

A scan can end failed; show the teacher that the check could not run and let them retry. Out of words returns 402.
Showing results fairly

Give teachers evidence, not verdicts

Show the sources

A plagiarism score shows matching text, not intent. Quotes, references and common phrases match too, so show each source and passage for the teacher to judge.

Show the sentences

Show which sentences were flagged and their likelihood, not only the overall AI score, so the teacher can discuss specific passages with the student.
No detector is perfect
An AI score is a likelihood, not proof that a student used AI. It can flag human writing — especially formal or formulaic text — and miss AI text that has been edited. Do not apply penalties automatically from a score.
FAQ

Plagiarism API for LMS: questions

Yes. The Plagiarism Checker Plus API is a REST API your LMS or edtech platform can call from its backend: send a submission’s text to POST /scans, poll GET /scans/{id}, and get a plagiarism score with sources plus an AI score with flagged sentences as JSON. You build the integration in your own code.

Build the integration on a free key

2,000 free API words and free test keys, no card, then pay as you go from $4.50 (50,000 words), or choose an API plan from $25 a month (API pricing). Running a school rather than a platform? See the plagiarism checker for institutions.

Canvas, Moodle and Blackboard are trademarks of their respective owners and are named only to say that we do not offer plugins for them. Use of the API is subject to our Terms of Service and Privacy Policy.