Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversaries may gain access to a system through a user visiting a website over the normal course of browsing.
.claude/skills/cyberstrikeus-t1456-drive-by-compromise/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 14% | 0% |
Adversaries may gain access to a system through a user visiting a website over the normal course of browsing. With this technique, the user's web browser is typically targeted for exploitation, but adversaries may also use compromised websites for non-exploitation behavior such as acquiring an Application Access Token.
Multiple ways of delivering exploit code to a browser exist, including:
Often the website used by an adversary is one visited by a specific community, such as government, a particular industry, or region, where the goal is to compromise a specific user or set of users based on a shared interest. This kind of targeted attack is referred to a strategic web compromise or watering hole attack. There are several known examples of this occurring.
Typical drive-by compromise process:
Platforms: Android, iOS
Determine if the target mobile environment is susceptible to Drive-By Compromise by examining the target platforms (Android, iOS).
Review whether mitigations for T1456 are in place. If defenses are absent or misconfigured, this technique may be exploitable.
Security updates frequently contain patches for known exploits.
| Finding | Severity | Impact | | ---------------------------------------- | -------- | -------------- | | Drive-By Compromise technique applicable | High | Initial Access |
| CWE ID | Title | | ------ | ------------------------- | | CWE-20 | Improper Input Validation |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 26,288 | 29,361 | +12% | 1 | 1 | 0% | 3,260 | 3,832 | +18% | 0 | 0 | — |
case-02 | fail→fail | 27,206 | 36,387 | +34% | 1 | 1 | 0% | 4,247 | 5,870 | +38% | 0 | 0 | — |
case-03 | fail→pass | 28,325 | 33,016 | +17% | 1 | 1 | 0% | 3,598 | 5,540 | +54% | 0 | 0 | — |
case-04 | pass→pass | 24,257 | 33,661 | +39% | 1 | 1 | 0% | 2,991 | 5,294 | +77% | 0 | 0 | — |
case-05 | pass→pass | 27,374 | 19,230 | -30% | 1 | 1 | 0% | 2,836 | 3,643 | +28% | 0 | 0 | — |
case-06 | pass→pass | 27,323 | 29,567 | +8% | 1 | 1 | 0% | 3,109 | 5,158 | +66% | 0 | 0 | — |
case-07 | pass→pass | 11,072 | 8,612 | -22% | 1 | 1 | 0% | 789 | 1,028 | +30% | 0 | 0 | — |
case-08 | fail→pass | 11,701 | 3,392 | -71% | 1 | 1 | 0% | 1,010 | 1,293 | +28% | 0 | 0 | — |
case-09 | fail→pass | 37,888 | 9,133 | -76% | 1 | 1 | 0% | 2,268 | 1,090 | -52% | 0 | 0 | — |
case-10 | pass→pass | 8,450 | 7,846 | -7% | 1 | 1 | 0% | 1,260 | 1,207 | -4% | 0 | 0 | — |
case-11 | pass→pass | 25,722 | 22,047 | -14% | 1 | 1 | 0% | 2,986 | 3,304 | +11% | 0 | 0 | — |
case-12 | pass→pass | 6,650 | 8,883 | +34% | 1 | 1 | 0% | 755 | 1,208 | +60% | 0 | 0 | — |
case-13 | pass→pass | 14,988 | 8,634 | -42% | 1 | 1 | 0% | 1,182 | 1,163 | -2% | 0 | 0 | — |
case-14 | pass→pass | 23,261 | 21,626 | -7% | 1 | 1 | 0% | 2,789 | 3,141 | +13% | 0 | 0 | — |
case-15 | pass→pass | 14,279 | 7,385 | -48% | 1 | 1 | 0% | 2,207 | 1,189 | -46% | 0 | 0 | — |
case-16 | fail→pass | 19,917 | 9,446 | -53% | 1 | 1 | 0% | 2,123 | 1,496 | -30% | 0 | 0 | — |
case-17 | fail→pass | 13,709 | 12,247 | -11% | 1 | 1 | 0% | 1,617 | 1,841 | +14% | 0 | 0 | — |
case-18 | pass→pass | 20,476 | 9,312 | -55% | 1 | 1 | 0% | 2,174 | 2,045 | -6% | 0 | 0 | — |
case-19 | fail→pass | 21,994 | 25,787 | +17% | 1 | 1 | 0% | 2,699 | 3,723 | +38% | 0 | 0 | — |
case-20 | pass→pass | 12,647 | 11,309 | -11% | 1 | 1 | 0% | 1,877 | 1,792 | -5% | 0 | 0 | — |
case-21 | pass→pass | 17,175 | 8,928 | -48% | 1 | 1 | 0% | 1,781 | 1,330 | -25% | 0 | 0 | — |
case-22 | pass→pass | 20,471 | 11,588 | -43% | 1 | 1 | 0% | 2,087 | 1,554 | -26% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +27 percentage points is the difference between those two pass rates over the 21 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.