What we will not do.
Roadmaps say what's coming. This is the other list: the things GenticFlow will not build and will not do while handling support tickets, even if they would close a deal.
Build an RMM.
GenticFlow monitors the environment and learns what normal looks like so it can spot emerging support issues. Our roadmap follows support outcomes: fewer tickets, verified resolutions, and less technician time per case. We are not pursuing RMM feature parity or building a third-party patching catalogue. Keep an existing RMM for the management functions you need, or use GenticFlow as your primary support platform where it fits.
Bypass the controls for an action.
Playbook and workflow commands follow their configured controls. Remote control, terminal, and API access have their own permissions and entitlements. API and MCP callers must enforce any required approval before requesting an action.
Train models on your data.
Customer data is not training data. If you need stricter isolation, bring your own model.
Treat redaction as a guarantee.
Redaction controls are designed to reduce sensitive values in model requests. Review your data and provider settings; these controls do not guarantee that every sensitive value is excluded.
Auto-execute destructive actions.
Volume deletes, drive formats, registry hive deletion, fork bombs, evidence destruction: blocked outright. Cannot be approved by anyone, including admins.
Bury the case history.
Users cannot edit or delete audit records through the application.
Feed a technician's private notes into model context.
Endpoint notes are a technician scratchpad. They are never used as model context and never shown to end users.
Lock you in with multi-year contracts.
Plans are month-to-month. We don't use legal terms as a sales tool. Annual is available if you prefer budget predictability.
AI inside the workflow
AI assists. Technicians own the outcome.
GenticFlow is a support platform with AI assistance inside the workflow. Daily support tools, useful evidence, and technician control remain central.
Answer user questions from your configured knowledge and retain the conversation for follow-up.
- Use end-user chat with your company knowledge.
- Carry the conversation into the support case.
Run read-only diagnostics, interpret the evidence, and explain findings and unresolved questions.
- Investigate an endpoint, a fleet, or a general question.
- Review findings and the sources behind them.
Run a supported playbook or continue with hands-on tools, under the applicable permissions and approval policies.
- Review proposed fixes and execute supported playbook actions.
- Use remote control, terminal, and files when hands-on work is needed.
Check whether the action addressed the issue and keep the evidence, outcome, and user updates with the case.
- Review available verification results and record the findings and actions.
- Turn repeated work into workflows with Hero.
AI helps your team move through the work. Your technicians own the support outcome.
Start with the daily support tools. Add AI assistance and configured automation where they are useful.
Action availability depends on your license, permissions, connected systems, and configured policies.
That is the bet.
The product worth building is the one that says no to the things that erode trust, so see it on a real ticket.