Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Process images using object detection, classification, and segmentation. Use when requesting "analyze image", "object detection", "image classification", or "computer vision". Trigger with relevant phrases based on skill purpose.
.claude/skills/dicklesworthstone-processing-computer-vision-tasks/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 48% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-15 | ✓→✗ | ▼ Worse | -10% | 0% |
This skill provides automated assistance for computer vision processor tasks.
This skill provides automated assistance for computer vision processor tasks. This skill empowers Claude to leverage the computer-vision-processor plugin to analyze images, detect objects, and extract meaningful information. It automates computer vision workflows, optimizes performance, and provides detailed insights based on image content.
/process-vision command, which processes the image and returns the results.This skill activates when you need to:
User request: "Analyze this image and identify all the cars and pedestrians."
The skill will:
User request: "Classify this image. Is it a cat or a dog?"
The skill will:
This skill utilizes the /process-vision command provided by the computer-vision-processor plugin. It can be integrated with other skills to further process the results of the computer vision analysis, such as generating reports or triggering actions based on detected objects.
The skill produces structured output relevant to the task.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 12,853 | 13,586 | +6% | 1 | 1 | 0% | 1,431 | 2,290 | +60% | 0 | 0 | — |
case-02 | fail→fail | 7,533 | 24,458 | +225% | 1 | 1 | 0% | 325 | 1,892 | +482% | 0 | 0 | — |
case-03 | fail→fail | 16,132 | 27,584 | +71% | 1 | 1 | 0% | 1,805 | 5,139 | +185% | 0 | 0 | — |
case-04 | fail→fail | 20,306 | 18,845 | -7% | 1 | 1 | 0% | 2,702 | 3,024 | +12% | 0 | 0 | — |
case-05 | pass→pass | 17,945 | 15,430 | -14% | 1 | 1 | 0% | 1,799 | 2,463 | +37% | 0 | 0 | — |
case-06 | pass→pass | 21,241 | 20,124 | -5% | 1 | 1 | 0% | 2,689 | 3,154 | +17% | 0 | 0 | — |
case-07 | pass→pass | 23,304 | 24,821 | +7% | 1 | 1 | 0% | 3,012 | 3,656 | +21% | 0 | 0 | — |
case-08 | pass→pass | 18,420 | 10,396 | -44% | 1 | 1 | 0% | 2,243 | 1,443 | -36% | 0 | 0 | — |
case-09 | pass→pass | 16,014 | 10,014 | -37% | 1 | 1 | 0% | 1,770 | 1,386 | -22% | 0 | 0 | — |
case-10 | pass→pass | 15,353 | 10,157 | -34% | 1 | 1 | 0% | 1,885 | 1,461 | -22% | 0 | 0 | — |
case-11 | fail→pass | 13,534 | 12,444 | -8% | 1 | 1 | 0% | 1,380 | 2,039 | +48% | 0 | 0 | — |
case-12 | fail→fail | 14,422 | 7,125 | -51% | 1 | 1 | 0% | 1,675 | 965 | -42% | 0 | 0 | — |
case-13 | fail→pass | 19,900 | 17,324 | -13% | 1 | 1 | 0% | 2,749 | 2,742 | -0% | 0 | 0 | — |
case-14 | fail→pass | 12,562 | 9,021 | -28% | 1 | 1 | 0% | 1,197 | 1,346 | +12% | 0 | 0 | — |
case-15 | pass→fail | 12,826 | 9,036 | -30% | 1 | 1 | 0% | 1,279 | 1,155 | -10% | 0 | 0 | — |
case-16 | pass→pass | 10,282 | 8,558 | -17% | 1 | 1 | 0% | 840 | 1,132 | +35% | 0 | 0 | — |
case-17 | fail→pass | 12,300 | 10,464 | -15% | 1 | 1 | 0% | 1,152 | 1,274 | +11% | 0 | 0 | — |
case-18 | pass→pass | 14,833 | 11,416 | -23% | 1 | 1 | 0% | 1,657 | 1,680 | +1% | 0 | 0 | — |
case-19 | pass→pass | 17,319 | 14,155 | -18% | 1 | 1 | 0% | 2,135 | 2,114 | -1% | 0 | 0 | — |
case-20 | pass→pass | 28,272 | 24,155 | -15% | 1 | 1 | 0% | 4,957 | 4,706 | -5% | 0 | 0 | — |
case-21 | pass→pass | 15,111 | 13,982 | -7% | 1 | 1 | 0% | 1,881 | 2,320 | +23% | 0 | 0 | — |
case-22 | pass→pass | 15,409 | 11,763 | -24% | 1 | 1 | 0% | 2,003 | 1,971 | -2% | 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 +14 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.