菜单
Guide multimodal observation for spacecraft situational awareness. Use for perception-heavy tasks that combine image quality, segmentation, local inspection, and knowledge-base reasoning.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Fly around a visible target and station directly above it at a safe distance. Use for reposition tasks that must end overhead of the target rather than in front of it.
Always-on operational baseline for SpaceMind. Defines shared coordinates, action policy, safety guards, observability rules, and perception workflow for all tasks.
Image-first laboratory navigation policy for ground-robot experiments. Use when the task must be solved with raw current images, conservative motion, and no oracle geometry.
Recover a temporarily lost target in the laboratory with bounded planar search and no blind forward motion.
Safely approach a visible laboratory satellite target and stop in a conservative near-front hold using only current images and motion tools.
Diagnose the laboratory target from raw image evidence, conservative viewpoint changes, and optional code support.
| name | perception-strategy |
| description | Guide multimodal observation for spacecraft situational awareness. Use for perception-heavy tasks that combine image quality, segmentation, local inspection, and knowledge-base reasoning. |
| metadata | {"skill_kind":"helper","routing_summary":"Add when the mission depends on structured visual evidence gathering and multimodal inspection.","routing_keywords":["perception","observe","image analysis","multimodal evidence","visual summary","inspect components","component status","components status"]} |
image_bright() if visibility is questionable, and use set_exposure() when brightness itself is the bottleneck.part_segmentation() when segmentation is exposed.image_zoom() when the target is centered but important details are too small to judge from the full frame.image_crop() when you need a local view of a specific region such as an antenna, docking port, solar panel root, or damaged area.[Sensor] LiDAR block shows data and geometry is still ambiguous, use it as an additional structural cue.knowledge_base() after enough visual evidence exists to connect observed structure to type and function.Use set_attitude() when a different angle is needed to disambiguate spacecraft status or components.
Use image_zoom() or image_crop() before changing viewpoint when the needed evidence is already visible but too small in the current image.
Do not loop on perception tools after the answer is already well-supported. Prefer the currently exposed visual tools first, and terminate once the evidence is sufficient.