Radical transparency, enterprise controls
Momentum measures activity — never content. Here is exactly what our desktop agents collect, what they never touch, and the isolation and governance controls wrapped around every tenant.
What we collect. What we never will.
Most monitoring tools bury this in a PDF. We put it on the front page: Momentum's agents capture activity signals and counts — never the content of anyone's work.
The desktop agent measures when a machine is actively used versus idle, against your configured workday and office hours.
Which applications and window titles are in the foreground — including offline apps — so classification and Timeline stay accurate.
Aggregate mouse-click and keystroke COUNTS as an activity signal. The agent counts events; it never records what was typed.
CPU and memory footprint of the agent itself, plus operational logs (Daemon, Installer, Task Scheduler, Upgrade) for supportability.
No keylogging, ever. Momentum records that keys were pressed — never which keys, and never text, passwords, or messages.
The current product has no screenshot or screen-recording capability. There is no toggle to switch it on.
The agent never accesses webcams, microphones, GPS, or any other physical sensor on the machine.
No document contents, email bodies, chat transcripts, or clipboard data are read, indexed, or transmitted.
One organization, one hard-walled tenant
Momentum is multi-tenant from the schema up. Every machine, user, setting, and activity record carries a tenant ID, and every query is scoped to exactly one. There is no shared pool an over-broad query could leak from — cross-tenant access simply doesn't exist as a code path.
Defense in depth, from agent to dashboard
Activity data is protected in motion, at rest, and at the moment of access — with the operational visibility to prove it.
Every machine earns its place — and can lose it instantly
Momentum treats the agent fleet like the security surface it is. Machines go through an explicit approval lifecycle before a single event is recorded, and admins can revoke a device remotely the moment it's lost, retired, or offboarded.
If the agent does it, there's a log line for it
Security teams shouldn't have to trust a black box on 500 endpoints. Momentum surfaces Info and Error logs across every agent component — Daemon, Installer, Task Scheduler, Upgrade — alongside per-machine CPU and memory usage, so IT can verify exactly what the agent is doing and what it costs.
Paperwork your security review actually needs
Certifications and commitments, ready to hand to procurement, your DPO, or a works council.
The questions your CISO will ask
Does Momentum capture screenshots or record screens?+
No. The current product has no screenshot or screen-recording capability at all — there is no hidden setting to enable it. Momentum's signal comes from active/idle time, app and window usage, and input counts.
Can admins read what employees type?+
No. The agent counts keystrokes and mouse clicks as an activity signal, but never records which keys were pressed. Keystroke content, passwords, and messages are never captured, transmitted, or stored.
How is one customer's data kept separate from another's?+
Every organization runs as an isolated tenant. All storage, queries, and API responses are scoped to a single tenant ID, so one tenant can never see another's data.
What happens when a laptop is lost or an employee leaves?+
Admins can remotely uninstall the agent or disable agent login for that machine immediately. Either action stops data collection from the device; disable-login preserves the install for reactivation.
Can any machine start reporting data as soon as the agent is installed?+
No. New machines register in a Pending state and report nothing until an Admin approves them. You can enable auto-approval for controlled, scripted rollouts — it's your call per tenant.
What compliance documentation can we get before buying?+
SOC 2 Type II report (under NDA), ISO 27001 certificate, our DPA with sub-processor list, data-residency options, and a security whitepaper describing the agent's exact collection behavior.