Security

• Every record is scoped to a workspace and enforced at the database layer, not only in the interface.
• Roles decide who can create or edit deals; other members hold read access.
• Privileged operations verify the caller's role on the server before they run.
• Report links shared outside the workspace use single-purpose tokens that owners can revoke.

• All traffic to the application and its APIs is encrypted in transit.
• Uploaded documents live in private storage readable only by the owning workspace.
• Managed database backups run on a regular schedule.
• Credentials and API keys are held in a secret store and never shipped to the browser.

• Server-side URL fetching validates targets to prevent requests to internal network addresses.
• Scheduled and webhook endpoints verify their caller before doing any work.
• Input to server functions is validated before use.
• Analyst actions such as score overrides are written to an append-only audit trail with a required reason.

The analysis layer is built to avoid fabricated certainty. Claims without a retrievable source are labelled as estimates or assumptions, conclusions carry a traceable path from data to evidence to reasoning to score, and analysis is gated when retrieval coverage is too thin to support a verdict.

If you believe you have found a vulnerability, contact us through our contact page with enough detail to reproduce it. Please avoid accessing other customers' data, degrading the service or publishing the issue before we have had a reasonable chance to fix it. We will acknowledge your report and keep you updated on the remediation.