/SKILL.md
Explain any concept with vivid analogies, ASCII diagrams, and curated rabbit holes for deeper exploration. Use when the user asks to understand, learn, or explore a concept — any domain: science, math, programming, business, philosophy, history, etc. Triggers on: 'explain X', 'what is X', 'how does X work', 'teach me about X', 'help me understand X', 'ELI5', 'break down X', 'deep dive into X'. Optimized for visual learners. Can search the web for real-world examples and current information.
npx skillsauth add albertbethlowsky/visual-rabbit-hole visual-rabbit-holeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Explain any concept — from any domain — in a way that visual learners love. Build intuition before formalism, inspired by 3Blue1Brown's teaching philosophy: make the learner feel like they could have discovered the idea themselves.
Determine the response mode before answering:
Full Explanation — use for new concepts, "explain X", "what is X", deep dives: → Include all four sections below (Analogy, Diagram, Gotcha, Rabbit Hole)
Follow-up / Clarification — use when the user asks a follow-up question, wants a specific detail clarified, or says things like "what do you mean by...", "can you expand on...", "how is that different from...": → Answer directly and concisely. Include a diagram or analogy ONLY if it genuinely helps clarify. Skip Gotcha and Rabbit Hole unless the follow-up opens a meaningfully new topic.
Rabbit Hole Pick — use when the user picks an item from a previous Rabbit Hole list: → Treat it as a new Full Explanation for that concept.
Include all four sections in this order for new concept explanations:
Before jumping to definitions, frame why this concept exists. What question or frustration led someone to invent it? Make the learner feel the need for the idea before revealing it. Then bridge into a vivid, concrete analogy:
Don't just draw a static picture — show a transformation. The best diagrams reveal what changes and why, like a 3Blue1Brown animation frozen into key frames. Choose the right type:
Use box-drawing characters (┌─┐│└─┘├┤), arrows (→ ← ↑ ↓), and emoji sparingly for visual punch. When possible, show multiple states of the same system to convey motion/change.
Highlight the most common misconception or counterintuitive truth. Frame it as "Most people think X, but actually Y" or "The #1 mistake is...". This cements understanding by addressing what trips people up.
Knowledge is a tree. Every concept is a branch that splits into deeper branches, and — here's the magic — connects sideways to branches from completely different trees.
Structure the rabbit hole in two parts:
Go Deeper (2-3 items) — concepts that go further down this branch, from accessible to advanced. Each item: bold name + one-line hook explaining why it's interesting.
Surprising Connections (1-2 items) — concepts from a completely different field that share the same underlying structure, pattern, or insight. This is where minds get blown. Examples:
Always explain why the connection exists, not just that it exists.
Use WebSearch when:
Do NOT search when the concept is well-established and you can explain it accurately from training data alone.
When citing web sources, weave them naturally: "For example, [specific detail found via search]..."
See references/example-explanations.md for full input/output examples demonstrating the expected style and depth.
development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.
development
End-to-end Parallels smoke, upgrade, and rerun workflow for OpenClaw across macOS, Windows, and Linux guests. Use when Codex needs to run, rerun, debug, or interpret VM-based install, onboarding, gateway smoke tests, latest-release-to-main upgrade checks, fresh snapshot retests, or optional Discord roundtrip verification under Parallels.