Skill
植物病害识别技能
Accurately identifies plant diseases based on computer vision and deep learning, supports both image and video input, outputs structured diagnostic reports i...
When to use 植物病害识别技能
Choose if
You're building an agricultural agent that needs to triage plant disease from field photos or short clips and you're willing to operate against the smyx cloud backend. Output is a structured Markdown report with disease classification, severity assessment, and prevention recommendations — convenient to forward to a human agronomist for verification.
Avoid if
You're not on the smyx cloud platform (the skill depends on its open-id flow and cloud-only historical queries), your inputs exceed 10MB, you need offline inference, or you'd treat the diagnosis as authoritative. The skill is bilingual (English/Chinese) and structured around the smyx framework — generic plant-CV alternatives exist if you need something framework-neutral.
Risk Flags
- HIGH scope SKILL.md is explicit that results are diagnostic references only, not medical/professional guidance, and require clear symptom visibility and proper lighting for accuracy. Do not use as the sole basis for crop-treatment decisions.
- MEDIUM privacy Like its smyx sibling skills, SKILL.md forbids reading local memory files; historical reports must be fetched from the smyx cloud. open-id acquisition follows a strict hierarchy (skill dir → workspace → user) — self-generation is not permitted. Temporary script generation is also prohibited.
- LOW scope Input file size capped at 10MB and limited to mp4/avi/mov video or jpg/jpeg/png images. Higher-resolution or longer captures will need pre-processing.
Cost
Type: Unknown
Distribution
- ClawHub
smyx-plant-disease-recognition-analysis- License
- MIT-0