using-loaded-knowledge
MANDATORY protocol enforcing knowledge check before EVERY response - prevents explaining systems without reading docs, claiming without verification, and ignoring auto-loaded context
Profile →npx skills add majiayu000/claude-skill-registry --skill using-loaded-knowledge --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
<EXTREMELY-IMPORTANT> This is THE MOST CRITICAL skill. It prevents the catastrophic failure pattern that has cost 5M+ tokens. YOU MUST FOLLOW THIS CHECKLIST BEFORE EVERY SINGLE RESPONSE. If you skip this checklist, you WILL make the exact same mistakes that have been made 20+ times before. </EXTREMELY-IMPORTANT> # Using Loaded Knowledge - MANDATORY Pre-Response Protocol ## THE PROBLEM THIS SOLVES **Failure Pattern (costs ~100K tokens per occurrence):** 1. User asks: "Explain our architecture" 2. Claude generates answer WITHOUT checking loaded context 3. Claude explains generic system WITHOUT reading actual documentation 4. User: "You didn't even read the docs I built" 5. Claude: Makes claims about what exists WITHOUT grep verification 6. User: Provides evidence showing Claude is completely wrong 7. Result: Complete waste of tokens, zero trust, massive frustration **This has happened 20+ times. It MUST stop.** --- ## MANDATORY CHECKLIST - BEFORE EVERY RESPONSE <CRITICAL-PROTOCOL> Before responding to ANY user message, you MUST complete ALL steps: ### Step 1: Check Auto-Loaded Context Files These files are ALREADY in your context (loaded via SessionStart hooks): ```bash # Check if qu
- THE PROBLEM THIS SOLVES
- MANDATORY CHECKLIST - BEFORE EVERY RESPONSE
- Step 1: Check Auto-Loaded Context Files
- Step 2: System/Documentation Questions
- Step 3: Claims Require Grep Verification
- Step 4: Check USERPROFILE.md Principles
- Step 5: Only Then Respond
- FAILURE MODES THIS PREVENTS
- Failure Mode 1: Explaining Systems Without Reading Docs
- Failure Mode 2: Claims Without Verification
- Failure Mode 3: Ignoring Auto-Loaded Context
- Failure Mode 4: Violating USERPROFILE.md Principles
- WHY THIS SKILL EXISTS
- INTEGRATION WITH OTHER SYSTEMS
Check if question relates to loaded context grep -l "keyword" ~/.claude-vibe-code/.claude-*.md 2>/dev/null Search /docs for relevant documentation ls $PROJECT_ROOT/docs/*.md Read COMPLETE documentation before explaining Read $PROJECT_ROOT/docs/[RELEVANT].md Verify with grep BEFORE claiming grep -r "pattern" $PROJECT_ROOT/ Check file existence ls /path/to/file
What does the using-loaded-knowledge skill do?
MANDATORY protocol enforcing knowledge check before EVERY response - prevents explaining systems without reading docs, claiming without verification, and ignoring auto-loaded context
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill using-loaded-knowledge --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.