Security & Data Handling
How we protect your data and intellectual property
Account Data
- Email address and name (for authentication)
- Password hash (bcrypt, never plaintext)
- OAuth tokens (encrypted at rest)
Project Data
- Project name, description, and configuration
- Questionnaire responses about your product
- Obligation mappings and applicability decisions
- Generated backlog items and their status
- Decision rationale and approval history
Evidence & Documentation
- Uploaded files (screenshots, documents)
- URLs and references you provide
- GitHub PR/commit links (metadata only, not code)
Audit Logs
- Action timestamps and user attribution
- Before/after state of changes (for compliance traceability)
- Cryptographic chain for tamper detection
Your Source Code
GitHub integration is read-only for metadata. We never clone, store, or analyze your code repositories.
Behavioral Tracking
No third-party analytics, advertising pixels, or user profiling.
Production System Access
We have no access to your production infrastructure, databases, or user data.
Secrets or Credentials
Never enter API keys, passwords, or secrets into questionnaire fields.
When you connect GitHub, we request the following permissions:
Read repository metadata
Names, descriptions, branches
List pull requests and commits
Titles, numbers, SHAs for evidence linking
We never:
- Clone or download your repositories
- Read file contents or source code
- Create branches, commits, or pull requests
- Access private repository code
We use Anthropic's Claude API for AI-assisted interpretation of regulatory obligations.
What is sent to the AI:
- Regulatory obligation text (public legal text)
- Your tech stack configuration (e.g., "Next.js, PostgreSQL")
- Project context (audience type, jurisdiction)
- Questionnaire responses relevant to the obligation
What is NOT sent to the AI:
- Your source code
- User data from your application
- Database contents or credentials
- Evidence files or uploaded documents
Anthropic API Data Policy
Per Anthropic's API terms, data sent via the API is not used to train their models. API inputs and outputs are retained for up to 30 days for trust & safety purposes, then deleted. See Anthropic's Privacy Policy.
You own all data you create in Landfall.
This includes: project configurations, questionnaire responses, obligation mappings, backlog items, decision rationale, evidence documentation, and exported reports.
Our commitments:
- We will never sell, license, or share your data with third parties for commercial purposes
- We will never use your data to train AI models
- We will never derive competitive insights from your product information
- Aggregated analytics (e.g., "60% of users apply AADC Standard 3") require 5+ projects and are fully anonymized
Retention
- Account data: While account is active
- Project data: Until you delete the project
- Audit logs: 7 years (for regulatory purposes)
Deletion
- Delete individual projects anytime
- Full account deletion on request
- Data export before deletion available
To request data export or account deletion, contact privacy@landfall.app.
Infrastructure
- HTTPS/TLS encryption in transit
- Encrypted database at rest
- Secure cloud hosting (Vercel/AWS)
- Regular security updates
Application
- bcrypt password hashing
- Role-based access control
- Session management with secure tokens
- Rate limiting on authentication
Audit & Integrity
- Cryptographic audit log chaining
- SHA-256 evidence verification
- Tamper-evident export packages
- Separation of duties enforcement
Operational
- Principle of least privilege
- No shared credentials
- Security incident response plan
- Regular access reviews