// TRUST CENTER

Your code is some of your most sensitive data.

Veyra is designed around minimizing access, isolating analysis, and giving you control over every connected asset. This page describes how that works, in enough detail that your engineers can evaluate it rather than take our word for it.

LAST REVIEWED: JANUARY 2026
// 01 · REPOSITORY ACCESS

Read-only, scoped, and revocable.

Veyra connects through a GitHub App you install on the specific repositories you choose. It requests the minimum permissions the analysis requires.

PermissionLevelWhy it is needed
Repository contentsReadRetrieve the code to be analyzed, including history for secret detection.
MetadataReadResolve repository names, default branches, and languages.
Pull requestsReadOnly where pull-request analysis is enabled on your plan.
Security eventsReadOnly where you choose to correlate existing GitHub security alerts.
Any write permissionNot requestedVeyra cannot modify, merge, or delete your source code.

Veyra cannot modify, merge, or delete your source code

No write permission is requested on any plan. If a future capability ever required one, it would be a separate installation prompt requiring your explicit, informed consent — never a silent scope expansion on an existing installation.

Credential handling

SHORT-LIVED TOKENS
Repository access uses short-lived installation tokens issued per assessment rather than long-lived personal access tokens.
SECRETS MANAGER
All credentials are held in a managed secrets store, never in source, configuration files, or environment templates committed to a repository.
NO PRODUCTION SECRETS IN SOURCE
Veyra applies the same rule to itself that it enforces for customers. Its own repositories are scanned for secrets on every push.
SCOPED PER TENANT
Credentials are scoped to a single customer organization. There is no shared credential with access across tenants.
// 02 · DATA ENCRYPTION

Encryption in transit and at rest.

IN TRANSIT
TLS 1.2 or higher for all external connections, including the dashboard, the API, source-control integrations, and webhook delivery. HTTP Strict Transport Security is enforced on all web properties.
AT REST
Findings, metadata, and any retained evidence are encrypted at rest using AES-256 through managed provider encryption.
BACKUPS
Backups are encrypted with the same standard, access-controlled separately from production, and restore procedures are tested.
SECRETS
Application secrets and integration credentials are held in a dedicated managed secrets store with access logged.
// 03 · ANALYSIS ISOLATION

Every assessment runs in a disposable environment.

The analysis environment exists for one assessment, for one customer, and is destroyed when the run completes.

01Customer authorizes repositories
02Control plane schedules assessment
03Ephemeral isolated scanner provisioned
04Short-lived read-only token issued
05Repository retrieved into sandbox
06Security analysis executes
07Normalized findings emitted
08Working copy destroyed
09Environment torn down
NO SHARED ANALYSIS STATE
Analysis environments are not reused between customers. Each assessment gets a clean environment.
TENANT ISOLATION
Customer data is logically isolated at the data layer, with tenant scoping enforced on every query — the same control Veyra checks for in your code.
ENVIRONMENT SEPARATION
Development, staging, and production are fully separate environments with separate credentials. Customer data is never copied into development or staging.
MINIMIZED AI EXPOSURE
Source code is not sent wholesale to third-party AI providers. Where AI assisted reasoning is applied, it operates on the minimum context required for the specific analysis.
// 04 · DATA RETENTION

What is kept, and for how long.

Veyra keeps the minimum needed to describe findings and show your risk trend over time.

DataRetentionNotes
Repository working copyDuration of the assessment onlyDestroyed when the analysis run completes.
Findings and metadata90 days / 1 year / 2 yearsBy plan: Essential, Growth, Advanced.
Code excerpts used as evidenceSame as the parent findingLimited to the lines needed to explain and reproduce the issue.
Audit logs1 year minimumSecurity-relevant actions, retained independently of plan.
Account and billing recordsAs required by lawPayment details are held by Stripe; Veyra does not store card numbers.
Backups35 days rollingDeleted data ages out of backups within this window.

Deletion means deletion

On request, Veyra deletes your findings, evidence, and repository metadata from production systems within 30 days, and from backups within the backup retention window. Audit records of security-relevant actions and records required for legal or accounting obligations are retained. You receive written confirmation when deletion completes.

// 05 · CUSTOMER CONTROL

You can withdraw access at any moment.

Control that requires a support ticket is not control.

IN THE DASHBOARD

  • Disconnect any repository immediately
  • Remove an application, domain, or API from scope
  • Pause all assessment activity
  • Adjust rate limits, excluded endpoints, and maintenance windows
  • Export your findings and reports
  • Request deletion of your organization’s data

IN GITHUB

You can revoke the Veyra GitHub App from your organization settings at any time, without notifying us. Access ends immediately and any assessment in progress fails closed.

Because Veyra never holds write access and never holds a long-lived token, revocation is complete rather than partial.

// 06 · ACCESS CONTROLS

Who at Veyra can see what.

MULTI-FACTOR AUTHENTICATION
Required for every Veyra team member on every system that touches customer data or infrastructure.
LEAST PRIVILEGE
Access is granted by role, scoped to what the role requires, and reviewed on a defined schedule and on every role change.
NO BLANKET SOURCE ACCESS FOR SUPPORT
Ordinary support personnel cannot browse customer source code. Support tooling exposes finding metadata, not repository contents.
JUST-IN-TIME ELEVATION
Where a specific investigation requires deeper access, it is time-bound, justified, approved, and logged.
ADMINISTRATIVE ACTIONS LOGGED
Every administrative action against a customer organization is recorded with actor, timestamp, and reason.
OFFBOARDING
Access is revoked as part of departure, and credentials the individual could have observed are rotated.
// 07 · AUDIT LOGGING

Security-relevant actions are recorded.

Both yours and ours. Audit records are available to your organization administrators.

LOGGED EVENTS

  • Authentication and session events
  • Repository connection and disconnection
  • Authorization agreement creation and change
  • Assessment start, completion, and failure
  • Finding status changes and severity adjustments
  • Report generation and export
  • User invitation, role change, and removal
  • Veyra administrative access to your organization
  • Data deletion requests and completion

Every record includes the actor, the affected resource, a timestamp, and where relevant the reason. Logs are retained for at least one year on every plan, independently of your finding-retention period, and are write-once from the perspective of the application.

Advanced plans include export of audit records for ingestion into your own logging or SIEM systems.

// 08 · SUBPROCESSORS

Third parties in the path of customer data.

This list is maintained as the platform develops. Customers are notified in advance of any material change.

SubprocessorPurposeData involved
Cloud infrastructure providerApplication hosting, databases, ephemeral analysis environmentsFindings, metadata, transient repository working copies
VercelMarketing site and form handlingContact form submissions
StripePayment processingBilling details. Veyra does not store card numbers.
Email delivery providerTransactional email and alertsEmail address, finding identifier and severity — never exploitable detail
GitHubSource control integrationRepository contents, accessed read-only under your installation

To be notified of subprocessor changes, email security@veyra.example and ask to be added to the notification list.

// 09 · CERTIFICATIONS

Where Veyra actually stands.

Veyra holds no security certifications at this time

Veyra is not SOC 2 certified, ISO 27001 certified, HIPAA compliant, PCI DSS certified, or FedRAMP authorized, and has not completed an independent third-party penetration test. Any vendor page claiming otherwise on our behalf is wrong. When these are genuinely achieved, they will be published here with the assessing body, the scope, and the date — and not one day earlier.

What Veyra can offer today is the technical detail on this page, a completed security questionnaire on request, and a direct conversation with the people who built the system. If your procurement process requires a certification Veyra does not hold, we will tell you that immediately rather than waste your evaluation cycle.

// 10 · DISCLOSURE

Found a problem in Veyra itself?

A company that asks for access to your source code should be prepared to receive security reports about its own systems, and should make that easy.

SECURITY CONTACT

security@veyra.example

Acknowledged within one business day. Safe harbour applies to good-faith research within the published scope. A machine-readable contact is published at /.well-known/security.txt.

Common questions

Can we review Veyra as a subprocessor for our own SOC 2?
Yes. This page covers most of what auditors ask for. For anything further, email security@veyra.example with your questionnaire and we will complete it directly rather than pointing you at a portal.
Where is our data stored?
Customer data is stored in the United States by default. If your contractual or regulatory position requires a specific region, raise it during scoping — we will tell you honestly whether we can accommodate it today.
What happens if Veyra has a security incident?
Veyra maintains an incident response plan covering detection, containment, investigation, and communication. Affected customers are notified without undue delay, with what is known, what is not yet known, and what action is required. Notification will not be delayed to make the disclosure more comfortable.
Do you have cyber liability insurance?
Coverage details are shared under NDA during procurement. Ask during scoping and we will give you the actual position.