How to Trick, Hack and Cheat Time Doctor 2 (2026)

How Time Doctor 2 measures your activity

Time Doctor 2's desktop agent records two things more or less continuously: keyboard and mouse input, and which application or website is in the foreground and for how long. The activity percentage you and your manager see is, at heart, the share of tracked time that contained any input at all — it is a coverage figure, not a measure of how hard you were working in any given moment.

Two things follow from that. A second with a single keystroke counts exactly the same as a second with thirty, so typing faster does not lift the number. And because reading a spec, watching a recorded call, or thinking through a problem generates almost no input, honest deep-focus work naturally scores lower than continuous data entry. The percentage is a presence signal, and it was never built to grade the quality of the work behind it.

Keyboard, mouse and the active window: what registers

Only two kinds of signal move the number: the keyboard and the mouse. Keystrokes count, mouse movement counts, clicks count, and scrolling counts. Time Doctor records that these events happened, not the words you typed — the input meter is measuring occurrence, not content. Alongside it, the agent logs which app was in front and, for browser work, the site or page, so the number sits next to a record of where your attention was pointed.

This is why the device carrying the input barely matters while the shape of it matters a great deal. A cursor drifting in one dead-straight line and a person nudging the trackpad while they read both produce input, but Time Doctor 2 does not stop at counting it — it also asks whether the pattern looks like a person, which is where a mechanical loop runs into trouble.

• Keystrokes — any key, regardless of what you type.
• Mouse movement, clicks and scrolling.
• The foreground app and, for browsers, the site or page — but not the content of your input.

The Unusual Activity Report, explained

The feature that separates Time Doctor 2 from a plain activity counter is its Unusual Activity Report. An automated check runs over recorded input, picks out sessions that do not look human, and surfaces them to the manager as a separate report — so a suspicious pattern is not buried in the raw numbers, it is escalated on its own.

Time Doctor describes the kinds of behaviour this check is built to catch, and several of them read like a description of a mouse jiggler. The exact thresholds are not published, and no outside guide can tell you precisely when a given account trips them, so treat the list below as the shape of what is flagged rather than a rulebook you can game to the millimetre.

• Unusually consistent, evenly spaced mouse clicks.
• Long mouse movement with no clicks at all.
• Uniform activity held for implausibly long stretches.

Why a mouse jiggler makes it worse, not better

A hardware dongle or a simple mouse-mover app repeats one motion forever: the same short path, at the same interval, with no clicks and no relationship to what is on screen. Under a plain activity counter that behaviour passes, because input is technically present. Under a repetition detector it is close to a signature — the very evenness that keeps the timer running is what marks the pattern as non-human.

So the trade is a bad one. A low activity percentage on its own is easy to explain: reading, meetings, a research day. An entry on the Unusual Activity Report is not, because it reframes the whole day as an attempt to fake input. A cheap jiggler converts a forgivable low number into a flagged one — the opposite of what its buyer wanted.

Screenshots, webcam and the idle timeout

The input meter is not the only channel. Depending on what your company enabled, Time Doctor can also take screenshots on a schedule, capture webcam shots, and enforce an idle timeout — after a set stretch with no input it stops the timer, and some setups show an "are you still working?" prompt. These run alongside the activity number, and a jiggler does nothing about most of them.

This matters because the channels are read together. A jiggler can keep the timer alive and hold the idle timeout off, but it cannot change a screen that has been frozen on the same view for an hour. A moving activity number sitting over a static screenshot is one of the clearest mismatches a reviewer can spot — far more telling than an ordinary quiet interval — because the two records are supposed to agree and they plainly do not.

What actually reads as a person

A real workday is uneven. Input arrives along varied cursor paths at irregular timing, mixes clicks and scrolling with keystrokes, dips for calls and reading, and climbs during hands-on work. Across the day it tends to settle into a natural band — very roughly 60–85% for input-driven roles — rather than pinning to any single value.

The pattern that draws attention is the opposite: a flat line at or near 100% for hours, or the identical reading in every interval. Nobody produces input in every sampled moment, so consistency at the very top of the scale reads as automation, not diligence. Variety and a believable middle range are what look human; a perfect, unbroken number is the real anomaly.

Practical, realistic guidance

If a machine is going to keep a session alive while you are away, the goal is not to maximise the number — it is to avoid the patterns the Unusual Activity Report is built to catch, and to keep the record internally consistent. That means variety over volume, and restraint over a wall of fake input.

None of this is magic, and it is worth being blunt about the limit: nothing can make low-input desk work look like high-input work while you are actually sitting there, and no tool can promise a specific tracker will never flag it. What is realistic is keeping a normal break from reading as a dead gap.

• Use it only while the computer is genuinely idle, not as a shift-long autopilot.
• Aim for a natural band (around 60–85%), not a flat 100%.
• Allow pauses — an unbroken line looks wrong to a repetition detector.
• Keep any activity inside apps you actually use, so it matches the screen.
• Do not try to disguise reading, meetings or a research day as heavy input.

Where Reaction fits

Reaction sits in one narrow place: the stretches when you step away but the session should stay alive. It runs only while the computer is genuinely idle, hands control back the instant you touch the mouse or keyboard, and produces varied, human-like input within a band you set — curved cursor paths, a living keyboard rhythm, occasional scrolling and window switches — rather than one motion on a loop. The point is to avoid the mechanical evenness a repetition detector like Time Doctor's is built to catch.

It does not, and cannot, make low-input work look like high-input work while you are actually at the desk doing it — that is a limit of the metric, not something to sell around. What it does is stop a normal break from reading as an absence. Use it within your agreements; the acceptable-use note below applies.