Security
Security and governance for your daily support platform.
GenticFlow brings tickets, endpoint context, and technician tools together. Review the access boundaries, action controls, audit records, and data handling for the capabilities your team uses.
Approval policies
Define what GenticFlow may run under policy and what needs designated approval. Configure per organization, playbook, or risk tier. Attached to automation workflows.
- Low risk actions (restart a service, clear temp files) can auto-execute
- Higher risk actions require technician or administrator sign-off, with single or multi-approver strategies
- Policies configurable per organization, action type, and risk level
Model safety guardrails
AI-assisted command execution uses risk checks and configured action controls. Direct technician tools and API actions have their own permissions and access requirements.
- Built-in playbook actions have an allow-list checked at execution time. Other command paths use their applicable permissions, risk checks, and approval configuration
- Three-tier risk classification (low, medium, critical) plus a fourth blocked tier. Each tier maps to a different execution path: auto-run under policy, route for approval, or refuse outright
- The blocked tier covers actions that cannot be safely executed in any context: filesystem destruction (rm -rf /, drive formatting), boot sector overwrites, registry hive deletion, BitLocker disable, fork bombs, and evidence destruction. These cannot be run and cannot be approved, by anyone, including admins
- Concrete examples of the medium tier: changing Windows Defender configuration, modifying HKLM registry keys, disabling a network interface, modifying firewall rules (iptables, pfctl, netsh advfirewall), restarting or shutting down a system, changing user passwords, modifying cron schedules, sudo package install/remove, and accessing sensitive files (shadow, sudoers, SSH keys). The applicable command checks and configured approval policy determine whether they are blocked or require approval
- Controls distinguish external content from instructions and limit available actions. They reduce risk but do not guarantee that model output or untrusted content is harmless
- Risky reversible actions (privilege escalation, force-kill of system processes, network configuration changes, package installation, scheduled task changes) are routed for approval through your approval policy. The assistant proposes with full context, and a designated approver signs off
- Pattern chaining detection: multiple low-risk patterns combined (download then decode then execute across separate lines) elevate the overall risk score even when no individual pattern is critical on its own
- Redaction controls are designed to reduce sensitive values in model requests. Review the data and provider settings for your deployment; these controls do not guarantee that every sensitive value is excluded
- Private technician notes on an endpoint are never used as model context and never shared with end users
- A model-provider failure can interrupt AI-assisted work. Review the investigation and command records before continuing; actions already performed are not automatically rolled back
- Every command is logged before and after execution. Command history traces a bad decision to the exact endpoint, command text, output, risk classification, and approver, so root-cause analysis is fast. Reversal of side effects is not automatic: rollback is the operator's call, informed by the logged before-state
Complete command history
Diagnostic runs, policy decisions, and approvals are all written down and available for compliance review.
- Diagnostic commands and outputs recorded verbatim
- Approval decisions with timestamp and approver identity
- Post-fix verification results
- Users cannot edit or delete audit records through the application
Bring your own model
Choose your model provider or connect a model hosted in your own environment. With a self-hosted model, model requests go to that configured endpoint. When you choose a hosted provider, that provider processes the requests under its terms and your configuration.
- OpenAI, Anthropic Claude, Google Gemini, any OpenAI API-compatible endpoint, or a self-hosted runtime via Ollama
- Per-tenant model selection: different organizations or workloads can use different models
- Self-hosted model endpoints are supported; confirm network and deployment requirements separately
- Bundled model routing can be fully disabled at the policy level when BYOM is required
Platform foundations.
Review the platform controls alongside the permissions for each support tool.
Identity and access
Two-factor authentication is mandatory for every account, by authenticator app (TOTP) or FIDO2 passkey, with no option to switch it off. Technicians can sign in through your identity provider over OpenID Connect: Microsoft Entra ID, Google Workspace, Okta, or any generic OIDC provider, with group-to-role mapping and account provisioning on first sign-in. Enforce single sign-on across the organization with a designated break-glass account, and set session timeouts to match your policy.
Trusted networks
Restrict who can reach the platform to the IP ranges you trust. Allow-lists apply separately to web and mobile sign-in, and access from outside the list is blocked before any data loads. Off until you configure it, with a built-in guard that stops you from locking yourself out when you enable it.
Tenant isolation
Each tenant has its own database credentials and its own object-storage credentials. Crossing the boundary between two tenants requires crossing a credential boundary, not just a code-level filter.
Encryption
TLS protects connections in transit, and HSTS directs browsers to use HTTPS. Sensitive fields are encrypted at the application layer, with per-tenant key separation.
Where your data lives
Default hosting is in Falkenstein, Germany. AI, payment, email, backup, and notification providers may process data in other locations listed in our DPA. Model-processing location depends on your configured provider. GenticFlow is not SOC 2 certified. Organizations with specific residency requirements can discuss deployment and provider options with us.
Your data, your exit
Command history, ticket history, and configuration are available in the platform, with API access where supported. Our DPA provides 30 days for export after the Service ends, deletion of Personal Data within 90 days, and retention of isolated backup copies for up to 180 days. Earlier deletion instructions and legal requirements may change these periods; the DPA sets out the applicable terms.
Signed binaries
Windows agents are signed with an Extended Validation Authenticode certificate. macOS agents are code-signed and notarised by Apple. You can verify the signature chain on every endpoint you install.
Operational commitments.
How we treat your data, how we respond when something goes wrong, and how we work with the security community.
Your data is not training data
GenticFlow does not use your data to train models. Managed model processing uses OpenAI's commercial API under terms that prohibit training on data submitted to the API. For stricter requirements, connect a model hosted in your own environment. Choosing a third-party model provider sends requests to that provider.
Incident response
For a Personal Data Breach, our DPA requires notice without undue delay and within 72 hours of awareness, or sooner where required. We do not wait for a completed investigation. Available information about the scope, likely consequences, and response is provided in phases as the investigation progresses.
Vulnerability disclosure
Security researchers and organizations can report suspected vulnerabilities to [email protected]. We acknowledge reports within 24 hours, work toward a fix on a 90-day disclosure timeline, and credit researchers who request it. Coordinated disclosure preferred.
For your security review.
Detailed security documentation is available under NDA as part of organization security reviews and vendor risk assessments. See our Data Processing Agreement for provider locations, retention periods, and processing terms.
Available on request:
- Security architecture and data-flow diagrams
- Information security policy and procedure summaries
- Subprocessor inventory and data processing agreement
- Incident response and breach notification processes
- Vulnerability management and patching cadence
- Vendor risk questionnaires (CAIQ, SIG Lite) on request
To start a security review, request vendor documentation, or report a vulnerability, write to [email protected]. We will respond with an NDA and the relevant documentation package.
See the support record yourself.
See user support, tickets, endpoint context, technician tools, and action controls in one support workflow.