Security
A watcher you can trust with real customer data.
Paperhuman is multi-tenant: every company runs on shared infrastructure, but no company can ever see another's data. That isolation is enforced automatically at the data-access layer, not by convention.
Per-company tenant isolation
Every database query is automatically scoped to the caller's company by the application's data-access layer, before it ever reaches Postgres.
Secrets encrypted at rest
Stored credentials, including SMTP/IMAP passwords, provider API keys, and OAuth tokens, are encrypted with AES-256-GCM before they touch the database.
Role-based access control
Viewer, agent, supervisor, admin. Assignment-scoped access is enforced server-side and visualized in a permission matrix, not just hidden UI.
Auditability
A full activity log records every action by actor. Deactivating a team member kills their session on the next request; password hashes never leave the API.
Enforced in code, not by convention
Every query is scoped before it ever reaches the database.
How tenant isolation actually works
Every request resolves to exactly one company before any data is read or written. From that point on, every query (reads, writes, updates, deletes, across every part of the product) passes through a single data-access layer that automatically attaches that company's id, with no route or feature able to opt out. That scoping is backed by a real foreign-key constraint at the database level, so a row can't exist without a valid company behind it. It's not an application-level filter you have to trust in isolation, it's enforced twice, at two different layers.
Let Paperhuman start watching today.
Connect your first channel or system in a few minutes, and only get pulled in when it matters.
Request a demo