{ "seed": 1601, "limitations": [ "Synthetic, small, English-heavy cases are not representative of production traffic.", "Contact labels cover parser false positives but do not establish consent, identity, or deliverability.", "Website fixtures classify supplied HTML only; they do not measure network, DNS, redirects, or adversarial pages.", "Latency and cache measurements are local pilot signals and vary by host load." ], "normalization": [ {"name": "SA contact fields", "input": {"name": " Acme Solar ", "website": "HTTPS://WWW.ACME.TEST/path", "phone": "082 555 1234", "email": " SALES@ACME.TEST "}, "expected": {"name": "Acme Solar", "website_domain": "acme.test", "email": "sales@acme.test"}}, {"name": "accented location", "input": {"name": "Cafe", "province": " Western Cape ", "city": "CAPE TOWN", "suburb": "Sea Point"}, "expected": {"province": "western cape", "city": "cape town", "suburb": "sea point"}} ], "matching": [ { "name": "shared exact domain is a positive", "source": {"name": "Acme Solar", "website": "https://www.acme.test", "city": "Johannesburg"}, "candidates": [ {"id": 1, "name": "Acme Solar (Pty) Ltd", "website": "http://acme.test/", "city": "Johannesburg"}, {"id": 2, "name": "Acme Supplies", "website": "https://supplies.test", "city": "Johannesburg"} ], "threshold": 0.72, "expected_ids": [1], "forbidden_ids": [2] }, { "name": "similar name without corroboration is not a positive", "source": {"name": "Brightline Consulting", "city": "Cape Town"}, "candidates": [ {"id": 3, "name": "Brightline Consulting", "city": "Durban"}, {"id": 4, "name": "Brightline Consulting", "city": "Cape Town"} ], "threshold": 0.72, "expected_ids": [4], "forbidden_ids": [3] }, { "name": "unrelated businesses are negatives", "source": {"name": "Green Oak Architects", "website": "green-oak.test"}, "candidates": [ {"id": 5, "name": "Green Oak Accounting", "website": "accounting.test"}, {"id": 6, "name": "Red River Bakery", "website": "redriver.test"} ], "threshold": 0.72, "expected_ids": [], "forbidden_ids": [5, 6] } ], "contacts": [ { "name": "public contacts and ignored markup", "source_url": "https://acme.test/contact", "html": "

Sales: sales@acme.test

Call +27 (12) 345-6789

API key: token@acme.test

", "expected": [["email", "sales@acme.test"], ["phone", "+27123456789"]], "forbidden": [["email", "secret@acme.test"], ["email", "tracker@tracker.test"], ["email", "token@acme.test"]] }, { "name": "forms are provenance not email destinations", "source_url": "https://bright.test/contact", "html": "

hello [at] bright.test

", "expected": [["email", "hello@bright.test"]], "forbidden": [["email", "email@bright.test"]] } ], "websites": [ {"name": "healthy", "status": 200, "url": "https://acme.test", "body": "

Acme Solar

Welcome to our business.

", "expected": "healthy"}, {"name": "parked", "status": 200, "url": "https://parked.test", "body": "Domain for sale - buy this domain", "expected": "parked"}, {"name": "construction", "status": 200, "url": "https://new.test", "body": "Website coming soon", "expected": "under_construction"}, {"name": "broken", "status": 404, "url": "https://gone.test", "body": "", "expected": "broken"}, {"name": "blocked", "status": null, "url": "https://private.test", "body": "", "error": "unsafe_address", "expected": "blocked"}, {"name": "unknown is not healthy", "status": null, "url": "https://unknown.test", "body": "", "expected": "unknown"} ], "scoring": [ {"name": "complete active business", "signals": {"business": {"name": "Acme", "email": "a@acme.test", "phone": "+27123456789"}, "website": {"classification": "healthy"}, "contacts": {"public_count": 1}, "domain": {"status": "resolved"}, "state": {"verified": true, "suppressed": false, "merge_status": "active"}}}, {"name": "suppressed remains ineligible", "signals": {"business": {"name": "Suppressed"}, "state": {"suppressed": true, "merge_status": "active"}}, "must_be_ineligible": true}, {"name": "stale evidence does not create positive", "signals": {"business": {"name": "Stale"}, "website": {"classification": "healthy", "stale": true}, "state": {"suppressed": false, "merge_status": "active"}}} ], "tenant_isolation": { "tenant_a": {"id": 101, "business_ids": [1, 2]}, "tenant_b": {"id": 202, "business_ids": [3, 4]}, "cross_tenant_ids": [3, 4] } }