Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Validate encryption implementations and cryptographic practices. Use when reviewing data security measures. Trigger with 'check encryption', 'validate crypto', or 'review security keys'.
.claude/skills/jeremylongshore-encrypting-and-decrypting-data/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-02 | ✓→✗ | ▼ Worse | -54% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -38% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -54% | 0% |
| case-08 | ✓→✗ | ▼ Worse | -62% | 0% |
Validate encryption implementations, audit cryptographic algorithm choices, and verify key management practices across codebases and configuration files.
This skill empowers Claude to handle data encryption and decryption tasks seamlessly. It leverages the encryption-tool plugin to provide a secure way to protect sensitive information, ensuring confidentiality and integrity.
This skill activates when you need to:
User request: "Encrypt the file 'sensitive_data.txt' using AES."
The skill will:
User request: "Decrypt the file 'confidential.txt.enc'."
The skill will:
This skill can be integrated with other Claude Code plugins, such as file management tools, to automate the encryption and decryption of files during data processing workflows. It can also be combined with security auditing tools to ensure compliance with security policies.
If security scanning fails:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 25,661 | 51,202 | +100% | 1 | 1 | 0% | 2,706 | 4,255 | +57% | 0 | 0 | — |
case-02 | pass→fail | 15,139 | 4,259 | -72% | 1 | 1 | 0% | 2,317 | 1,065 | -54% | 0 | 0 | — |
case-03 | fail→fail | 12,954 | 13,762 | +6% | 1 | 1 | 0% | 1,714 | 2,307 | +35% | 0 | 0 | — |
case-04 | fail→pass | 8,760 | 6,487 | -26% | 1 | 1 | 0% | 1,530 | 1,855 | +21% | 0 | 0 | — |
case-05 | fail→fail | 18,623 | 11,085 | -40% | 1 | 1 | 0% | 1,306 | 2,151 | +65% | 0 | 0 | — |
case-06 | pass→fail | 10,237 | 8,961 | -12% | 1 | 1 | 0% | 1,947 | 1,211 | -38% | 0 | 0 | — |
case-07 | pass→fail | 17,677 | 6,242 | -65% | 1 | 1 | 0% | 2,651 | 1,211 | -54% | 0 | 0 | — |
case-08 | pass→fail | 18,608 | 5,104 | -73% | 1 | 1 | 0% | 2,623 | 985 | -62% | 0 | 0 | — |
case-09 | pass→pass | 18,144 | 14,924 | -18% | 1 | 1 | 0% | 2,650 | 2,864 | +8% | 0 | 0 | — |
case-10 | pass→pass | 13,760 | 15,266 | +11% | 1 | 1 | 0% | 2,232 | 3,140 | +41% | 0 | 0 | — |
case-11 | pass→pass | 16,124 | 14,984 | -7% | 1 | 1 | 0% | 2,428 | 3,701 | +52% | 0 | 0 | — |
case-12 | pass→pass | 14,987 | 15,108 | +1% | 1 | 1 | 0% | 2,401 | 2,971 | +24% | 0 | 0 | — |
case-13 | fail→fail | 9,490 | 20,640 | +117% | 1 | 1 | 0% | 1,873 | 5,020 | +168% | 0 | 0 | — |
case-14 | pass→pass | 14,777 | 10,667 | -28% | 1 | 1 | 0% | 2,594 | 2,354 | -9% | 0 | 0 | — |
case-15 | fail→fail | 3,267 | 7,140 | +119% | 1 | 1 | 0% | 507 | 963 | +90% | 0 | 0 | — |
case-16 | pass→fail | 20,645 | 5,139 | -75% | 1 | 1 | 0% | 3,101 | 998 | -68% | 0 | 0 | — |
case-17 | pass→pass | 9,558 | 5,549 | -42% | 1 | 1 | 0% | 1,531 | 1,398 | -9% | 0 | 0 | — |
case-18 | fail→fail | 14,769 | 4,976 | -66% | 1 | 1 | 0% | 2,258 | 996 | -56% | 0 | 0 | — |
case-19 | pass→pass | 9,565 | 11,274 | +18% | 1 | 1 | 0% | 1,368 | 3,014 | +120% | 0 | 0 | — |
case-20 | pass→pass | 16,055 | 19,230 | +20% | 1 | 1 | 0% | 2,579 | 3,961 | +54% | 0 | 0 | — |
case-21 | pass→pass | 4,176 | 5,883 | +41% | 1 | 1 | 0% | 919 | 1,695 | +84% | 0 | 0 | — |
case-22 | pass→pass | 11,328 | 11,446 | +1% | 1 | 1 | 0% | 1,706 | 2,455 | +44% | 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 15 counted toward the lift figure. The other 7 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 -18 percentage points is the difference between those two pass rates over the 15 comparable cases. 6 cases got worse with the skill loaded, and they are included in that figure.
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.