AC mediapipe
Use Google MediaPipe for on-device computer vision — face, hand, and pose landmark detection, face detection, object detection, and image segmentation on images, videos, or webcam streams. Activate when the user asks to detect faces/hands/poses/objects, extract landmarks or keypoints, do background removal or selfie segmentation, or analyze photos/videos/webcam with MediaPipe. Provides scripts/run_mediapipe.py supporting both the legacy solutions API (bundled models) and the Tasks API (model file required), plus landmark index references. | 使用 Google MediaPipe 进行端侧计算机视觉：人脸/手势/姿态关键点检测、人脸检测、目标检测和图像分割，支持图片、视频和摄像头。当用户要求检测人脸/手/姿态/目标、提取关键点、背景移除（自拍分割）或使用 MediaPipe 处理图片/视频/摄像头时激活。提供 scripts/run_mediapipe.py，同时支持传统 solutions API（内置模型）与 Tasks API（需模型文件），并附带关键点索引参考。
As a process C 54/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 54/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 1 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 21 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2125 tokens
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +3Description length 765: enough signal without eating the budget
- +4Structure: 14 headings
- +3Step-by-step instructions: 21 items
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
- +3All 1 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 91.