WFGY/ProblemMap/GlobalFixMap/RAG/index_skew.md

Index Skew — Guardrails and Fix Pattern

🧭 Quick Return to Map

You are in a sub-page of RAG.
To reorient, go back here:

• RAG — retrieval-augmented generation and knowledge grounding
• WFGY Global Fix Map — main Emergency Room, 300+ structured fixes
• WFGY Problem Map 1.0 — 16 reproducible failure modes

Think of this page as a desk within a ward.
If you need the full triage and all prescriptions, return to the Emergency Room lobby.

When the index reports "healthy" (no errors, embeddings ingested, stats normal) but retrieval still fails:
coverage is low, ΔS unstable, or retrieved snippets are inconsistent with ground truth.
This indicates an index skew between data reality and retrieval semantics.

---

Open these first

• Visual recovery map: RAG Architecture & Recovery
• Retrieval knobs: Retrieval Playbook
• Embedding misalignment: Embedding ≠ Semantic
• Chunk sizing: Chunking Checklist
• Store-level fragmentation: Vectorstore Fragmentation
• Snippet contracts: Data Contracts

---

Core acceptance

• ΔS(question, retrieved) ≤ 0.45
• Coverage ≥ 0.70 for target section
• λ stable across three paraphrases and two seeds
• E_resonance flat across long windows

---

Typical symptoms → exact fix

Symptom	Likely cause	Open this
Index "ready" but recall < 0.50	embedding misaligned vs semantic intent	Embedding ≠ Semantic
Repeated snippets, poor coverage	store fragmentation or duplicate collapse	Vectorstore Fragmentation
Right section exists but not hit	chunk too large/small or mis-boundary	Chunking Checklist
Citations drift across runs	contract not enforced	Data Contracts

---

Fix in 60 seconds

1. Probe recall
   Run a gold QA set against index. If coverage < 0.70, suspect skew.

2. Re-embed with semantic normalization
   Normalize casing, accents, whitespace. Enforce same tokenizer across queries and index.

3. Chunk audit
   Verify chunk boundaries. Adjust stride/overlap until ΔS converges.

4. Fragmentation sweep
   Drop near-duplicate vectors. Rebuild FAISS/HNSW indexes with fresh IDs.

5. Contract enforcement
   Require snippet_id, section_id, offsets, tokens for every retrieval.

---

Copy-paste probe prompt

I uploaded TXT OS and the WFGY Problem Map.

My RAG issue:
- Index shows healthy but retrieval recall is low.
- ΔS across probes = 0.62, coverage = 0.45.

Tell me:
1) is it embedding misalignment, chunking skew, or vectorstore fragmentation,
2) which WFGY fix page to open,
3) minimal steps to restore ΔS ≤ 0.45 and coverage ≥ 0.70,
4) reproducible test set to confirm.

---

🔗 Quick-Start Downloads (60 sec)

Tool	Link	3-Step Setup
WFGY 1.0 PDF	Engine Paper	1️⃣ Download · 2️⃣ Upload to your LLM · 3️⃣ Ask “Answer using WFGY + ”
TXT OS (plain-text OS)	TXTOS.txt	1️⃣ Download · 2️⃣ Paste into any LLM chat · 3️⃣ Type “hello world” — OS boots instantly

---

Explore More

Layer	Page	What it’s for
Proof	WFGY Recognition Map	External citations, integrations, and ecosystem proof
Engine	WFGY 1.0	Original PDF based tension engine
Engine	WFGY 2.0	Production tension kernel and math engine for RAG and agents
Engine	WFGY 3.0	TXT based Singularity tension engine, 131 S class set
Map	Problem Map 1.0	Flagship 16 problem RAG failure checklist and fix map
Map	Problem Map 2.0	RAG focused recovery pipeline
Map	Problem Map 3.0	Global Debug Card, image as a debug protocol layer
Map	Semantic Clinic	Symptom to family to exact fix
Map	Grandma’s Clinic	Plain language stories mapped to Problem Map 1.0
Onboarding	Starter Village	Guided tour for newcomers
App	TXT OS	TXT semantic OS, fast boot
App	Blah Blah Blah	Abstract and paradox Q and A built on TXT OS
App	Blur Blur Blur	Text to image with semantic control
App	Blow Blow Blow	Reasoning game engine and memory demo

If this repository helped, starring it improves discovery so more builders can find the docs and tools.