Paste the description text, upload a PDF or Word document, or enter the URL of the posting. All three feed into the same analysis engine.
For files, we extract raw text using a PDF parser or Word document reader. For URLs, we fetch and scrape the page. All text is cleaned and standardised before analysis.
Our engine scans the text against a library of weighted rules — each one corresponding to a known scam pattern. Each matching rule adds to the risk score; each positive signal (professional email domain, detailed requirements) adds to the legitimacy score.
An AI model (XGBoost + BERT) analyses the cleaned text for subtler signals the rules alone might miss. The AI score is blended with the rule score using confidence-weighted averaging.
The result shows your risk score (0–100), legitimacy score, every specific red flag found, positive signals, a plain-English explanation, and recommended next steps tailored to the verdict.
The risk score runs from 0 (no indicators) to 100 (maximum indicators). Here's what each band means:
Multiple serious indicators detected. Do not proceed. Do not pay or share any personal or financial information.
Several warning signs present. Research the company thoroughly and verify the contact independently before responding.
Few or no major red flags. Still do your own due diligence and confirm contact details through official channels.
Every analysis runs these checks, each weighted by how strongly it correlates with fraud:
Copy any job description — from WhatsApp, email, a website, or anywhere — and paste it directly. Optionally add the job title, company, and source for a more accurate analysis.
Upload a PDF or Word document (.pdf, .doc, .docx) up to 10 MB. We extract the text automatically and run the full analysis on the content.
Submit the URL of a job posting. We fetch the page, extract content, analyse the domain and URL patterns against known scam domains, and check for the canonical employer source.
Paste a job description and get your risk score in seconds.
Analyse a job posting