Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversaries may enumerate files and directories or search in specific device locations for desired information within a filesystem.
.claude/skills/cyberstrikeus-t1420-file-and-directory-discovery/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -67% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -56% | 0% |
Adversaries may enumerate files and directories or search in specific device locations for desired information within a filesystem. Adversaries may use the information from File and Directory Discovery during automated discovery to shape follow-on behaviors, including deciding if the adversary should fully infect the target and/or attempt specific actions.
On Android, Linux file permissions and SELinux policies typically stringently restrict what can be accessed by apps without taking advantage of a privilege escalation exploit. The contents of the external storage directory are generally visible, which could present concerns if sensitive data is inappropriately stored there. iOS's security architecture generally restricts the ability to perform any type of File and Directory Discovery without use of escalated privileges.
Platforms: Android, iOS
Determine if the target mobile environment is susceptible to File and Directory Discovery by examining the target platforms (Android, iOS).
Review whether mitigations for T1420 are in place. If defenses are absent or misconfigured, this technique may be exploitable.
Security architecture improvements in each new version of Android and iOS make it more difficult to escalate privileges. Additionally, newer versions of Android have strengthened the sandboxing applied to applications, restricting their ability to enumerate file system contents.
| Finding | Severity | Impact | | ------------------------------------------------- | -------- | --------- | | File and Directory Discovery technique applicable | High | Discovery |
| CWE ID | Title | | ------- | --------------------------------- | | CWE-200 | Exposure of Sensitive Information |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | fail→fail | 25,916 | 25,995 | +0% | 1 | 1 | 0% | 3,060 | 3,624 | +18% | 0 | 0 | — |
case-01 | fail→pass | 33,431 | 41,759 | +25% | 1 | 1 | 0% | 4,422 | 3,685 | -17% | 0 | 0 | — |
case-02 | fail→fail | 34,904 | 25,997 | -26% | 1 | 1 | 0% | 4,055 | 3,055 | -25% | 0 | 0 | — |
case-03 | fail→fail | 31,351 | 39,491 | +26% | 1 | 1 | 0% | 4,077 | 5,289 | +30% | 0 | 0 | — |
case-04 | fail→pass | 21,050 | 16,039 | -24% | 1 | 1 | 0% | 2,477 | 2,393 | -3% | 0 | 0 | — |
case-05 | fail→pass | 21,257 | 12,409 | -42% | 1 | 1 | 0% | 2,873 | 961 | -67% | 0 | 0 | — |
case-06 | fail→pass | 21,775 | 16,027 | -26% | 1 | 1 | 0% | 2,302 | 1,010 | -56% | 0 | 0 | — |
case-07 | pass→pass | 16,656 | 15,736 | -6% | 1 | 1 | 0% | 1,636 | 2,354 | +44% | 0 | 0 | — |
case-08 | pass→pass | 17,830 | 21,784 | +22% | 1 | 1 | 0% | 2,697 | 2,814 | +4% | 0 | 0 | — |
case-09 | fail→pass | 17,323 | 6,881 | -60% | 1 | 1 | 0% | 1,877 | 830 | -56% | 0 | 0 | — |
case-10 | pass→pass | 10,174 | 13,483 | +33% | 1 | 1 | 0% | 1,439 | 1,700 | +18% | 0 | 0 | — |
case-11 | pass→pass | 34,257 | 18,578 | -46% | 1 | 1 | 0% | 2,381 | 2,681 | +13% | 0 | 0 | — |
case-12 | fail→pass | 23,524 | 10,358 | -56% | 1 | 1 | 0% | 2,530 | 1,434 | -43% | 0 | 0 | — |
case-13 | pass→pass | 12,022 | 7,212 | -40% | 1 | 1 | 0% | 1,662 | 791 | -52% | 0 | 0 | — |
case-14 | pass→pass | 23,371 | 30,458 | +30% | 1 | 1 | 0% | 3,129 | 4,068 | +30% | 0 | 0 | — |
case-15 | fail→pass | 21,556 | 3,638 | -83% | 1 | 1 | 0% | 2,315 | 1,146 | -50% | 0 | 0 | — |
case-16 | pass→pass | 19,807 | 24,964 | +26% | 1 | 1 | 0% | 2,671 | 3,271 | +22% | 0 | 0 | — |
case-21 | pass→pass | 27,525 | 27,718 | +1% | 1 | 1 | 0% | 3,383 | 3,664 | +8% | 0 | 0 | — |
case-17 | pass→pass | 13,141 | 8,924 | -32% | 1 | 1 | 0% | 1,742 | 1,000 | -43% | 0 | 0 | — |
case-18 | pass→pass | 20,014 | 9,079 | -55% | 1 | 1 | 0% | 2,278 | 1,748 | -23% | 0 | 0 | — |
case-19 | pass→pass | 26,823 | 23,991 | -11% | 1 | 1 | 0% | 2,280 | 3,360 | +47% | 0 | 0 | — |
case-20 | pass→pass | 35,089 | 27,609 | -21% | 1 | 1 | 0% | 1,947 | 2,650 | +36% | 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. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 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.