# π₯ Semantic Clinic Index
**A complete triage hub for AI failures β beyond the core 16 β powered by WFGY.**
Use this page when you donβt yet know *which thing is breaking*. Start from symptoms, jump to a failure family, then open the exact fix page. All fixes are driven by WFGY instruments: `ΞS` (semantic stress), `Ξ»_observe` (layered observability), and `E_resonance` (coherence control).
> If this page saves you time, a β helps others find it.
---
## How to use this page
1. **Identify the symptom** in the table below.
2. **Open the family** (Prompting / Retrieval / Reasoning / Memory / Agents / Infra / Eval).
3. **Follow the fix page**, then verify with ΞS β€ 0.45 and convergent Ξ».
Prefer a pipeline-first view (OCR β chunk β embed β store β retrieve β prompt β LLM)?
Read **[`RAG Architecture & Recovery`](./rag-architecture-and-recovery.md)**.
---
Want to explore all failure types or test fixes directly?
β See the full **[Problem Map Index](./README.md)** for categorized issues and fixes.
---
## Quick triage by symptom
| Symptom you see | Likely family | Open this |
|---|---|---|
| Answers cite wrong snippet / mismatch with ground truth | Retrieval β RAG | [`hallucination.md`](./hallucination.md) |
| Chunks look right but reasoning is wrong | Reasoning | [`retrieval-collapse.md`](./retrieval-collapse.md) |
| High similarity, wrong meaning | Retrieval / Embeddings | [`embedding-vs-semantic.md`](./embedding-vs-semantic.md) |
| Model canβt explain *why* (no trace) | Observability | [`retrieval-traceability.md`](./retrieval-traceability.md) |
| Output collapses over long dialogs / 100 k tokens | Memory / Long-context | [`long-context-stress.md`](./long-context-stress.md) |
| Jailbreak or prompt injection succeeds | Prompting / Safety | [`prompt-injection.md`](./prompt-injection.md) |
| Vector store index βlooks fineβ but retrieval irrelevant | Retrieval / Data | [`vectorstore-metrics-and-faiss-pitfalls.md`](./vectorstore-metrics-and-faiss-pitfalls.md) |
| OCR PDFs **look** correct yet answers drift | Data / OCR | [`ocr-parsing-checklist.md`](./ocr-parsing-checklist.md) |
| Multi-agent tools fight each other | Agents | [`multi-agent-chaos.md`](./multi-agent-chaos.md) |
| First prod call crashes after deploy | Infra / Boot | [`predeploy-collapse.md`](./predeploy-collapse.md) |
> Still lost? Open the **Beginner Guide** symptom checklist first.
---
## Families & maps (with exact fixes)
### A) Prompting & Safety
Guard against injections, jailbreaks, role drift, and schema leakage.
- **Prompt Injection** β [`prompt-injection.md`](./prompt-injection.md)
- **System Prompt Drift** β [`system-prompt-drift.md`](./system-prompt-drift.md)
- **Citation-first, schema-locked prompting** β [`retrieval-traceability.md`](./retrieval-traceability.md)
- **Overconfidence / Bluffing Controls** β [`bluffing.md`](./bluffing.md)
- **Safety Boundary Map (overview)** β [`Safety_Boundary_Problems.md`](./Safety_Boundary_Problems.md)
**Verification**: ΞS(question, context) β€ 0.45; Ξ» remains convergent across paraphrases; injection probes do not flip Ξ».
---
### B) Retrieval, Data & Vector Stores
Make the index correct, measured, and explainable.
- **Hallucination & Chunk Drift** β [`hallucination.md`](./hallucination.md)
- **Interpretation vs Retrieval Collapse** β [`retrieval-collapse.md`](./retrieval-collapse.md)
- **Embedding β Semantic Meaning** β [`embedding-vs-semantic.md`](./embedding-vs-semantic.md)
- **Traceability (why this snippet?)** β [`retrieval-traceability.md`](./retrieval-traceability.md)
- **Vector Store Metrics & FAISS Pitfalls** β [`vectorstore-metrics-and-faiss-pitfalls.md`](./vectorstore-metrics-and-faiss-pitfalls.md)
- **Semantic Chunking Checklist** β [`chunking-checklist.md`](./chunking-checklist.md)
- **OCR / Parsing Quality Gate** β [`ocr-parsing-checklist.md`](./ocr-parsing-checklist.md)
**Verification**: coverage β₯ 0.70 to target section; ΞS(question, retrieved) β€ 0.45; flat-high ΞS vs k β index/metric mismatch.
---
### C) Reasoning & Logic Control
Detect and repair logic collapse, dead ends, and abstraction failures.
- **Logic Collapse & Recovery** β [`logic-collapse.md`](./logic-collapse.md)
- **Long Reasoning Chains** β [`context-drift.md`](./context-drift.md)
- **Symbolic Collapse** β [`symbolic-collapse.md`](./symbolic-collapse.md)
- **Philosophical Recursion** β [`philosophical-recursion.md`](./philosophical-recursion.md)
- **Reasoning Schemas (cite β explain, bridge nodes)** β [`reasoning-schemas.md`](./reasoning-schemas.md)
- **Tool Router Debug** β [`tool-router-debug.md`](./tool-router-debug.md)
**Verification**: fix point when Ξ» stays convergent after applying BBCR (bridge) + BBAM (variance clamp).
---
### D) Memory & Long-Context
Keep threads coherent across sessions and very long windows.
- **Memory Breaks Across Sessions** β [`memory-coherence.md`](./memory-coherence.md)
- **Entropy Collapse** β [`entropy-collapse.md`](./entropy-collapse.md)
- **Long-Context Stress Map** β [`long-context-stress.md`](./long-context-stress.md)
- **Conversation Stitching & Memory Design** β [`memory-design-patterns.md`](./memory-design-patterns.md)
**Verification**: E_resonance flat; ΞS stable at window joins.
---
### E) Multi-Agent & Orchestration
Coordinate tools, roles, and shared memory without conflict.
- **Multi-Agent Chaos** β [`multi-agent-chaos.md`](./multi-agent-chaos.md)
- **Agent Boundary Design** β [`agent-boundary-design.md`](./agent-boundary-design.md)
- **Consensus Protocols for Agents** β [`agent-consensus-protocols.md`](./agent-consensus-protocols.md)
**Verification**: when agents couple, ΞS does not spike; arbitration logs traceable.
---
### F) Infra / Deploy
Boot in a known-good order, every time.
- **Bootstrap Ordering** β [`bootstrap-ordering.md`](./bootstrap-ordering.md)
- **Deployment Deadlock** β [`deployment-deadlock.md`](./deployment-deadlock.md)
- **Pre-Deploy Collapse** β [`predeploy-collapse.md`](./predeploy-collapse.md)
- **Observability Runbook** β [`observability-runbook.md`](./observability-runbook.md)
**Verification**: deterministic warm-up; idempotent index builds; version/secret checks pass.
---
### G) Evaluation & Guardrails
Detect βdouble hallucinationβ and prevent regression.
- **Evaluation Playbook** β [`evaluation-playbook.md`](./evaluation-playbook.md)
- **WFGY Metrics Spec** β [`wfgy-metrics.md`](./wfgy-metrics.md)
**Acceptance**: retrieve QA coverage β₯ 0.70 and ΞS β€ 0.45; Ξ» convergent; repeatability across seeds.
---
## Ask the AI to fix your AI (safe prompt)
```txt
Read the WFGY TXT OS and Problem Map docs. Extract ΞS, Ξ»_observe, E_resonance and the modules (BBMC, BBPF, BBCR, BBAM).
Given my failure:
- symptom: [describe]
- traces: [ΞS probes, Ξ» states if any]
Tell me:
1) which layer/family is failing and why,
2) which fix page to open,
3) minimal steps to push ΞS β€ 0.45 and keep Ξ» convergent,
4) how to verify the fix.
````
---
### π§ Explore More
| Module | Description | Link |
|-----------------------|----------------------------------------------------------|----------|
| Problem Map 1.0 | Initial 16-mode diagnostic and symbolic fix framework | [View β](https://github.com/onestardao/WFGY/edit/main/ProblemMap/README.md) |
| Problem Map 2.0 | RAG-focused failure tree, modular fixes, and pipelines | [View β](https://github.com/onestardao/WFGY/blob/main/ProblemMap/rag-architecture-and-recovery.md) |
| Semantic Clinic Index | Expanded failure catalog: prompt injection, memory bugs, logic drift | [View β](./SemanticClinicIndex.md) |
| Semantic Blueprint | Layer-based symbolic reasoning & semantic modulations | [View β](https://github.com/onestardao/WFGY/tree/main/SemanticBlueprint/README.md) |
| Benchmark vs GPT-5 | Stress test GPT-5 with full WFGY reasoning suite | [View β](https://github.com/onestardao/WFGY/tree/main/benchmarks/benchmark-vs-gpt5/README.md) |
---
> π **Early Stargazers: [See the Hall of Fame](https://github.com/onestardao/WFGY/tree/main/stargazers)** β
> Engineers, hackers, and open source builders who supported WFGY from day one.
> β Help reach 10,000 stars by 2025-09-01 to unlock Engine 2.0 for everyone β Star WFGY on GitHub