detect_object
100 / 100 · Low risk
Fine to try.
Low-level open-vocabulary object detection + segmentation via Grounding-DINO + SAM (Rollout Tier 2 perception). Pass a noun phrase ("red cube", "hammer head", "silver bowl"), get back bounding box + mask centroid + detection confidence. NO VLM in the loop — fully deterministic. THIS SKILL IS THE PERCEPTION CORE: the real Grounding-DINO + SAM implementation (model singleton, per-frame cache, `detect()` pure function, `Detection` dataclass) lives in this skill's policy.py. Every other perception…
What this skill can do
We found nothing in the files we read that matches our rules. That does not prove the skill is safe: see the limits of static scanning.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'detect-object' ] || [ -L ~/.agents/skills/'detect-object' ]; then echo 'A skill folder named detect-object already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/nssmd/RoboRSI' "$d" && git -C "$d" checkout --quiet '90c8569705bf4004a4f79e34e650c87ab7f075d4' -- 'roborsi/embodied/skills/base/detect_object/robotwin' && mkdir -p ~/.agents/skills && cp -R "$d"/'roborsi/embodied/skills/base/detect_object/robotwin' ~/.agents/skills/'detect-object' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'detect-object' ~/.claude/skills/'detect-object' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
[](https://zalavijaysinh.github.io/skilltrust/repos/nssmd/RoboRSI/)
Score history
| Date | Version | Score |
|---|---|---|
| 2026-11-13 | 90c8569 | 100 |