visual_diff
100 / 100 · Low risk
Fine to try.
Before/after frame diff for plan-react-replan. Captures a frame as "before", optionally executes an action via a follow-up call, then captures "after" and asks the VLM: "what changed? did the intended action happen?". Composes a side-by-side panel (before | after | pixel diff heatmap) so the VLM has all three signals in one image. Replaces "VLM stares at after-frame alone trying to infer if it matches intent".
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/'visual-diff' ] || [ -L ~/.agents/skills/'visual-diff' ]; then echo 'A skill folder named visual-diff 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 "$d" checkout --quiet '90c8569705bf4004a4f79e34e650c87ab7f075d4' -- 'roborsi/embodied/skills/base/visual_diff/robotwin' && mkdir -p ~/.agents/skills && cp -R "$d"/'roborsi/embodied/skills/base/visual_diff/robotwin' ~/.agents/skills/'visual-diff' && 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/'visual-diff' ~/.claude/skills/'visual-diff' (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 |