Security at Lot & Block
Lot & Block holds regulated housing data, so this page does two things most vendors won't: it says exactly what we do today, and it says plainly what we haven't done yet.
Tenant isolation, and how it's tested
Every record in Lot & Block carries the organization it belongs to, and every API route resolves that organization from the authenticated session — never from anything the client sends.
We don't ask you to take that on faith: automated cross-tenant tests run on every code change. An HTTP isolation gate exercises our API routes as a second organization and fails the build if any route reveals or touches another organization's data, alongside cross-tenant write specs and a database isolation suite.
How we use AI
AI never makes a final compliance determination. It drafts, extracts, summarizes and flags; a person reviews and signs off before anything regulated leaves the building. Drafts carry citations and confidence flags so the reviewer can check the work.
AI output is schema-validated before the application accepts it, and our AI model provider's commercial terms prohibit training on customer data.
Access and authentication
Sign-in is handled by Clerk, with multi-factor authentication available to every user. Access to customer data by our own operators through internal tooling is limited per organization, rate limited, and recorded in an append-only audit log.
Code and change security
Every code change is scanned in CI for vulnerable dependencies, insecure patterns and leaked secrets (dependency audit, Semgrep and gitleaks).
Data handling
Customer data is encrypted in transit (TLS) and at rest. Documents live in private, organization-scoped storage with server-derived paths, and downloads go through short-lived signed URLs — never public links.
What we haven't done yet
- No SOC 2 report yet. The founder has taken a prior company through SOC 2 Type 1, and we are building the policy set and evidence trail now — but we won't say "SOC 2" until a report is issued.
- No third-party penetration test yet. One is planned once we have revenue to fund it properly.
- We don't yet publish retention, backup-restore (RPO/RTO), uptime SLA or breach-notification commitments. Those arrive with our business-continuity and incident-response policies, not before.
Security questions, or something on this page that doesn't match what you see? Email us — we'd rather correct it than let it stand.