Developer documentation
Tradelynx Property Intelligence API
Consent-floored, verified-at-source property intelligence. Every response is evidence to support human decisions — never the sole basis for an adverse decision affecting a person. Access is commercial: an API key with the right scope, plus a per-organisation entitlement and feature flag, are all required (contact enterprise@tradelynx.co.uk). Sandbox keys (tlx_test_) return deterministic synthetic data and need no entitlement.
OpenAPI 3 spec (JSON) → · enterprise@tradelynx.co.uk
Authentication & access
Send your key as Authorization: Bearer tlx_live_…. Access to each product needs three things: a key with the right scope, a per-organisation entitlement (commercial), and the product's feature flag enabled for your organisation — so nothing is reachable by accident. Keys are issued in your workspace under API keys.
Sandbox: issue a tlx_test_… key and call the same endpoints — you get deterministic synthetic data (same id → same fixture), no entitlement needed, and no path to real data. Build first, contract second.
Endpoints
One call returns identity + evidence-graded statutory-compliance posture + a consented minimised history summary for a property in YOUR portfolio (foreign ids return 404). Gates: property:read scope + property_api entitlement + enterprise.property_api flag.
Up to 50 property_ids per call; page through larger books. Partial results: a slot not in your portfolio returns ok:false in place without failing the batch. Same gates as the single endpoint; consent is checked per property.
The org-wide compliance calendar (CHG-2026-0560). Gates: compliance_api entitlement + enterprise.compliance_api flag + compliance-scoped key.
Insurer organisations only; requires risk:read scope + risk_api entitlement + flag, and the property owner's standing risk_pricing consent. See your commercial agreement.
For a marketplace property_id (body: { property_id }), returns the underwriting profile over the property owner's CONSENTED work history for a property your insurer org is CONNECTED to — verified servicing cadence per trade + water-event mentions, aggregated only (never a derived score). consent:false when the owner has no standing risk_pricing consent. Requires underwriting:read scope + risk_api entitlement + flag. Evidence for a human underwriter, never an automated decision.
For a marketplace property_id, returns claim-relevant evidence over the consented, connected work history (was the boiler serviced? prior water damage? wiring recent?) with provenance. Evidence only — a human decides. Requires claims:read scope + risk_api entitlement + flag.
For an address (body: { address: { line1?, postcode } }), returns the EPC band + improvement pathway (official EPC register) and the flood peril band (Environment Agency open data). PUBLIC reference data — no consent required, no personal data. MEES F/G flagged as current law; a proposed EPC-C-by-2030 rental minimum is flagged as exposure (consultation, not yet law). Intelligence, not a lending decision. Requires lender:read scope + lender_api entitlement + enterprise.lender_api flag.
Webhooks (monitoring)
Move from point-in-time queries to continuous monitoring: subscribe an endpoint in your workspace under Webhooks and receive signed, retried deliveries. Payloads are deliberately thin — { property_id, changed_at } — you pull the anchor endpoint for detail, so consent and audit are enforced on every read and the push channel can never leak.
Thin payload — { property_id, changed_at } only. Fetch the anchor endpoint for detail (consent + audit are enforced on the pull, never bypassed by the push). Configure endpoints in your workspace under Webhooks; deliveries are signed and retried.
Thin payload — { property_id, changed_at } only. Emitted only while the owner's disclosure consent stands.
The standing notice
Every response carries it, and it is a condition of use: Tradelynx supplies evidence to support human decisions — this data must never be the sole basis for an adverse decision affecting a person. History data is only ever disclosed under the property owner's standing, revocable consent, and every read is audit-logged.