Data Handling

Where Your Data Lives

RequirementsHub builds a Knowledge Model from your documents — a structured map of stakeholders, objectives, systems, constraints, and requirements. This page explains how that data is stored, encrypted, and deleted, and which third parties process it. Last Updated: June 2026.

What Is the Knowledge Model?

When you upload a document, we extract entities — named stakeholders, objectives, systems, constraints, requirements, and the relationships between them. This structured model is what powers gap detection, traceability, and maturity scoring. The original document is preserved alongside it.

Data Storage & Encryption

Data TypeWhere StoredEncrypted
Original documentsSupabase Storage (AWS S3)Yes (AES-256)
Document chunksSupabase database (PostgreSQL)Yes (AES-256)
Knowledge entitiesSupabase database (PostgreSQL)Yes (AES-256)
Gap findingsSupabase database (PostgreSQL)Yes (AES-256)
Generated outputsSupabase database (PostgreSQL)Yes (AES-256)
Stripe payment dataStripe (not stored by us)Yes (per Stripe policy)

All data in transit is protected with TLS 1.3.

Third-Party Processors

ServicePurposeData Shared
SupabaseDatabase and file storageAll project data
AnthropicAI entity extraction and analysis (primary)Document text chunks
OpenAIAI generation, repair, and fallback (secondary)Knowledge Model context
Google Gemini (via Lovable AI Gateway)AI fallback (tertiary, failover only)Knowledge Model context
StripePayment processingEmail, payment details
MailgunTransactional emailEmail address, email content
CloudflareApplication runtime and edge routingRequest metadata

Anthropic, OpenAI, and Google are accessed under API terms that prohibit using customer data for model training. The full subprocessor list with purpose and region is at /trust/subprocessors.

Retention Schedule

Data CategoryActive RetentionPost-Deletion Purge (Active Systems)Backup RetentionPost-Cancellation GraceLog Retention
Workspace content (requirements, decisions, objectives, stakeholders, comments, activity)Retained for the life of the workspacePurged from active systems within 30 days of deletion requestRemoved from backups within 90 days30 days read-only grace, then deletionn/a
Account & profile data (email, name, auth identifier)Retained while account is activePurged from active systems within 30 days of account deletionRemoved from backups within 90 days30 days grace before account deletionn/a
Workspace membership recordsRetained while workspace existsPurged within 30 days of workspace deletion or member removalRemoved from backups within 90 daysFollows workspace lifecyclen/a
Uploaded documents (X-Ray source files)Retained while project existsPurged from object storage within 30 days of project deletionRemoved from backups within 90 daysFollows project lifecyclen/a
AI analysis outputs (gap findings, scores, generated artifacts)Retained while project existsPurged within 30 days of project deletionRemoved from backups within 90 daysFollows project lifecyclen/a
Billing & payment records (invoices, subscription state)Retained for 7 years after issue (financial recordkeeping)Not purged on account deletion — retained as required by lawPer Stripe policyn/an/a
Stripe payment data (card details, payment methods)Stored only by Stripe; not stored by RequirementsHubPer Stripe customer deletion processPer Stripe policyPer Stripe policyn/a
Access & security logs (authentication, admin actions, security events)Retained for security monitoring and incident investigationNot purged on account deletion — required for security audit trailIncluded in standard backup cyclen/a12 months
Email communications (transactional emails sent via Mailgun)Retained by Mailgun per their retention policyPer Mailgun policy on requestPer Mailgun policyn/aPer Mailgun policy
Invitation records (pending and accepted workspace invites)Pending invitations retained until accepted, revoked, or expired (30 days)Purged within 30 days of workspace deletionRemoved from backups within 90 daysFollows workspace lifecyclen/a
  • "Active systems" means the production database and object storage that serve live application requests. Once data is purged from active systems it is no longer accessible through the application or APIs.
  • Backup retention reflects the rolling window during which deleted data may remain in encrypted backups before being overwritten. Backups are not restored or accessed except during disaster recovery.
  • To request deletion outside the self-service flow, email privacy@requirementshub.ai. We will confirm completion in writing.

Data Residency

Customer data at rest is hosted in Supabase on AWS us-east-1 (Northern Virginia). We do not currently offer tenant-level region selection. Enterprise customers with data residency requirements should contact sales@requirementshub.ai.

Questions

Last updated: June 2026