Fake-Candidates-Checker An Apify Actor for checking candidate authenticity. It validates provided candidate data, crawls linked evidence (profiles, resumes, employer verification URLs), and produces a scored verdict with anomalies and evidence links. ## Project structure - .actor/actor.json — Actor configuration (name, version, runtime, metadata) - .actor/input_schema.json — Candidate-focused input (name, emails, phones, claimed employers, profile/resume URLs, crawler limits) - .actor/output_schema.json — Links to dataset view with verdicts - .actor/dataset_schema.json — Output view definition for verdicts, anomalies, and evidence - src/main.js — Actor entry point (Cheerio-based crawler with validation, scoring, and evidence aggregation) - Dockerfile — Container definition for deployment ## What it checks - Candidate identity presence on provided pages (name match) - Claimed employer mentions on provided/verifier URLs - Email domains against employer domains (when provided) - Phone number format sanity checks - Aggregated evidence with per-URL status, title, and hits - Simple scoring (0–100) and verdict buckets: verified, needs_review, high_risk ## Development workflow 1. Install dependencies (skip if already installed in your environment): bash npm install 2. Run the Actor locally: bash apify run 3. Authenticate with Apify (once per environment): bash apify login 4. Deploy to the Apify platform: bash apify push ## Notes - Uses CheerioCrawler (no Playwright) for speed; add Playwright only if you need JS rendering. - Respect robots.txt and site ToS; provide only URLs you are permitted to fetch. - The storage/ directory is ignored locally and created automatically by Apify during runs.