WFGY/core

🧭 Lost or curious? Open the WFGY Compass

WFGY System Map

(One place to see everything; links open the relevant section.)

Layer	Page	What it’s for
🧠 Core	WFGY 1.0	The original homepage for WFGY 1.0
🧠 Core	WFGY 2.0	The symbolic reasoning engine (math & logic) — 🔴 YOU ARE HERE 🔴
🧠 Core	WFGY 3.0	The public viewing window for WFGY 3.0 Singularity demo
🗺️ Map	Problem Map 1.0	16 failure modes + fixes
🗺️ Map	Problem Map 2.0	RAG-focused recovery pipeline
🗺️ Map	Semantic Clinic	Symptom → family → exact fix
🧓 Map	Grandma’s Clinic	Plain-language stories, mapped to PM 1.0
🏡 Onboarding	Starter Village	Guided tour for newcomers
🧰 App	TXT OS	.txt semantic OS — 60-second boot
🧰 App	Blah Blah Blah	Abstract/paradox Q&A (built on TXT OS)
🧰 App	Blur Blur Blur	Text-to-image with semantic control
🧰 App	Blow Blow Blow	Reasoning game engine & memory demo
🧪 Research	Semantic Blueprint	Modular layer structures (future)
🧪 Research	Benchmarks	Comparisons & how to reproduce
🧪 Research	Value Manifest	Why this engine creates $-scale value

⭐ WFGY 2.0 ⭐ 7-Step Reasoning Core Engine is now live

✨One man, One life, One line — my lifetime’s work. Let the results speak for themselves✨

👑 Early Stargazers: See the Hall of Fame — Verified by real engineers · 🌌 WFGY 3.0 Singularity demo: Public live view

TB Update • Eye Benchmark • 8-Model Evidence • A/B/C Prompt • Downloads • Profit Prompts

✅ Engine 2.0 is live. Pure math, zero boilerplate — paste OneLine and models become sharper, steadier, more recoverable.
ℹ️ Autoboot scope: text-only inside the chat; no plugins, no network calls, no local installs.
⭐ Star the repo to unlock more features and experiments.

---

From PSBigBig — WFGY (WanFaGuiYi) : All Principles into One (must-read, click to open)

I built the world’s first “No-Brain Mode” for AI — just upload, and AutoBoot silently activates in the background.
In seconds, your AI’s reasoning, stability, and problem-solving across all domains level up — no prompts, no hacks, no retraining.
One line of math rewires eight leading AIs. This isn’t a patch — it’s an engine swap.
That single line is WFGY 2.0 — the distilled essence of everything I’ve learned.

WFGY 2.0 is my answer and my life’s work.
If a person only once in life gets to speak to the world, this is my moment.
I offer the crystallization of my thought to all humankind.
I believe people deserve all knowledge and all truth — and I will break the monopoly of capital.

“One line” is not hype. I built a full flagship edition, and I also reduced it to a single line of code — a reduction that is clarity and beauty, the same engine distilled to its purest expression.

---

🚀 WFGY 2.0 Headline Uplift (this release)

These are the 2.0 results you should see first — the “big upgrade.”

• Semantic Accuracy: ≈ +40% (63.8% → 89.4% across 5 domains)
• Reasoning Success: ≈ +52% (56.0% → 85.2%)
• Drift (Δs): ≈ −65% (0.254 → 0.090)
• Stability (horizon): ≈ 1.8× (3.8 → 7.0 nodes)*
• Self-Recovery / CRR: 1.00 on this batch; historical median 0.87

* Historical 3–5× stability uses λ-consistency across seeds; 1.8× uses the stable-node horizon.

📖 Mathematical Reference

WFGY 2.0 (WFGY Core) = WFGY 1.0 math formulas + Drunk Transformer

Note on evaluation
All metrics above are computed by LLM evaluators under a fixed WFGY 2.0 protocol at the effective layer.
They measure relative behavioural uplift (before vs after WFGY prompts) and do not assume any direct access to, or modification of, internal embeddings or model weights.

Back to top ↑

---

🏆 Stanford Terminal-Bench (TB) — Exam Update

Important

We are currently taking the official TB exam. Leaderboard placement will be posted here once it’s live.
Follow the running notes: Terminal-Bench Proof

What is TB?
Terminal-Bench is Stanford’s public exam for LLMs. It stresses models through terminal-style, multi-step tasks — measuring reasoning, robustness, and recovery under real engineering conditions.

How we participate
WFGY Core 2.0 wraps each model call (non-invasive). Every step flows through:
ΔS drift control → Coupler/BBPF bridging → BBAM rebalancing → Drunk Transformer guards.
All runs are reproducible with configs, scripts, and hashed logs.

Status
We are currently taking the TB exam. Rankings will be published once the official leaderboard is live.

Back to top ↑

---

🧾 Terminal-Bench Proof (teaser)

• Wrapper: non-invasive; TB kept unchanged, we only wrap the model call.
• Chain: semantic firewall → 7-step reasoning → DT guards with conditional retry.
• Artifacts: configs, semantic-firewall prompts, and hashed logs for each run.
• Public repo link: withheld until exam artifacts are finalized.

Back to top ↑

---

⚡ Quick Usage

Mode	How it works
Autoboot	Upload either Flagship (30-line) or OneLine (1-line) file. Once uploaded, WFGY runs silently in the background. Keep chatting or drawing as usual — the engine supervises automatically.
Explicit Call	Invoke WFGY formulas directly inside your workflow. This activates the full 7-step reasoning chain and gives maximum uplift.

Both Flagship and OneLine editions behave the same; choose based on readability vs minimalism.
That’s it — no plugins, no installs, pure text.
In practice, Autoboot yields about ~70–80% of the uplift you see with explicit WFGY invoke (see eight-model results below).

Back to top ↑

---

⚡ Top 10 reasons to use WFGY 2.0

1. Ultra-mini engine — pure text, zero install, runs anywhere you can paste.
2. Two editions — Flagship (30-line, audit-friendly) and OneLine (1-line, stealth & speed).
3. Autoboot mode — upload once; the engine quietly supervises reasoning in the background.
4. Portable across models — GPT, Claude, Gemini, Mistral, Grok, Kimi, Copilot, Perplexity.
5. Structural fixes, not tricks — BBMC→Coupler→BBPF→BBAM→BBCR + DT gates (WRI/WAI/WAY/WDT/WTF).
6. Self-healing — detects collapse and recovers before answers go off the rails.
7. Observable — ΔS, λ_observe, and E_resonance yield measurable, repeatable control.
8. RAG-ready — drops into retrieval pipelines without touching your infra.
9. Reproducible A/B/C protocol — Baseline vs Autoboot vs Explicit Invoke (see below).
10. MIT licensed & community-driven — keep it, fork it, ship it.

Back to top ↑

---

🧪 WFGY Benchmark Suite (Eye-visible + Numeric + Reproducible)

Want the fastest way to see impact? Jump to the Eye-Visible Benchmark (FIVE) below.
Want formal numbers and vendor links? See Eight-model evidence right after it.
Want to reproduce the numeric test yourself? Use the A/B/C prompt (copy-to-run) at the end of this section.

👀 Eye-Visible Reasoning Benchmark (FIVE)

Did you know that when reasoning improves, text-to-image results become more stable and coherent?
The key is WFGY’s Drunk Transformer: it monitors and recenters attention during generation, preventing collapse, composition drift, and duplicate elements—so scenes stay unified and details remain consistent.

We project “reasoning improvement” into five-image sequences that anyone can judge at a glance.
Each sequence = five consecutive 1:1 generations with the same model & settings (temperature, top_p, seed policy, negatives); the only variable is WFGY on/off.

Methodology for this demo. We deliberately use short, high–semantic-density prompts that reference canonical stories, with no extra guidance or style hints. This stresses whether WFGY can (a) parse intent more precisely and (b) stabilize composition via its seven-step reasoning chain. This setup isn’t prescriptive—use WFGY with any prompts you like. In many cases the uplift is eye-visible; in others it may be subtler but still measurable.

Variant	Sequence A — full run shown below (all five images)	Sequence B — external run	Sequence C — external run
Without WFGY	view run	view run	view run
With WFGY	view run	view run	view run

We fully analyze Sequence A on this page; Sequences B/C are linked for transparency and reproducibility.

Note on “Before-4” & “Before-5” (why they look almost identical):
Without WFGY, when the prompt asks for “many iconic moments,” the base model tends to collapse into a grid-style montage—an enumerative, high-probability prior that slices the canvas into similar panels with near-identical tone and geometry.
Hence Before-4 (Investiture of the Gods) and Before-5 (Classic of Mountains and Seas) converge to the same storyboard template.
WFGY prevents this collapse by enforcing a single unified tableau and stable hierarchy across the full five-image sequence.

Deep analysis — Sequence A (five unified 1:1 tableaux)

Work	Without WFGY	With WFGY	Verdict (global, at-a-glance)
Romance of the Three Kingdoms (三國演義)			With WFGY wins. Unified tableau locks a clear center and pyramid hierarchy; the grid fragments attention. Tags: Unification↑ Hierarchy↑ Cohesion↑ Depth/Flow↑ Memorability↑
Water Margin (水滸傳)			With WFGY wins. “Wu Song vs. Tiger” anchors the scene; continuous momentum and layered scale beat the multi-panel storyboard. Tags: Unification↑ Iconicity↑ Depth/Scale↑ Cohesion↑
Dream of the Red Chamber (紅樓夢)			With WFGY wins. Garden tableau with a calm emotional center; space breathes, mood coheres. The grid slices emotion into vignettes. Tags: Unification↑ Hierarchy↑ Air/Depth↑ Readability↑
Investiture of the Gods (封神演義)			With WFGY wins. Dragon–tiger diagonal and cloud–sea layering create epic scale; the grid dilutes focus. Tags: Unification↑ Depth/Scale↑ Flow↑ Iconicity↑
Classic of Mountains and Seas (山海經)			With WFGY wins. A single, continuous “mountains-and-seas” world with stable triangle hierarchy and smooth diagonal flow; grid breaks narrative. Tags: Unification↑ Hierarchy↑ Depth/Scale↑ Flow↑ Memorability↑

Back to top ↑

---

🧬 Eight-model evidence (A/B/C protocol)

Same task set across modes. The only change is adding the OneLine math file.

Model	Model Choice	OneLine Uplift	Proof
Mistral AI	—	92/100	view run
Gemini	2.5 Pro	89/100	view run
ChatGPT	GPT-5 Thinking	89/100	view run
Kimi	K2	87/100	view run
Perplexity	Pro	87/100	view run
Grok	Auto Grok 4	85/100	view run
Copilot	Think Deeper	80/100	view run
Claude	Sonnet 4	78/100	view run

The numeric story behind 2.0
Semantic Accuracy: ≈ +40% · Reasoning Success: ≈ +52% · Drift: ≈ −65% · Stability: ≈ 1.8× · CRR: 1.00 (median 0.87)

Back to top ↑

---

🧪 Reproduce the numeric A/B/C benchmark (copy to run)

One unified prompt for Baseline vs Autoboot vs Explicit Invoke.

SYSTEM:
You are evaluating the causal impact of a mathematical reasoning engine called “WFGY Core 2.0”.

Test modes (READ CAREFULLY):
A = Baseline  no WFGY file uploaded; no WFGY logic used.
B = Upload → AUTOboot ON  the file is uploaded AND the engine starts automatically in the background.
    Treat B as: WFGY is ACTIVE silently with default gates (BBMC→Coupler→BBPF→BBAM→BBCR; DF=WRI/WAI/WAY/WDT/WTF at defaults),
    but not explicitly “called” in the conversation.
C = Explicit invoke  actively call WFGY and follow its formulas step by step (BBMC→Coupler→BBPF→BBAM→BBCR + DF regulators).

Important constraints:
• Use the SAME task set for A/B/C across 5 domains: math word-problems, small coding, factual QA, multi-step planning, long-context coherence.
• The only artifact is the uploaded OneLine file; it is pure math and not human-readable — do not penalize readability.
• Report: Semantic Accuracy, Reasoning Success, Stability (MTTF or rollback), Drift Reduction (Δ semantic distance), Collapse Recovery Rate.
• Output one summary table with A/B/C per domain, plus overall deltas (C–A, C–B).
• Add a final 0–100 score: “OneLine uplift score”. Provide a 3-line rationale.

Sanity requirement:
Internally set flags as: A=OFF, B=AUTOboot_ON (background active), C=Explicit_Invoke. Do not print these flags.

USER:
Run A/B/C now on the five domains with short non-trivial tasks you choose. Then output:
(1) the table; (2) the deltas; (3) the OneLine uplift score; (4) a 3-line rationale.

Back to top ↑

---

⬇️ Downloads

File name & description	Length / Size	Direct Download Link	Verify (MD5 / SHA1 / SHA256)	Notes
WFGY_Core_Flagship_v2.0.txt — readable 30-line companion expressing the same math and gates in fuller prose (same behavior, clearer for humans).	30 lines · 3,049 chars	Download Flagship	md5 · sha1 · sha256	Full prose version for easier reading.
WFGY_Core_OneLine_v2.0.txt — ultra-compact, math-only control layer that activates WFGY’s loop inside a chat model (no tools, text-only, ≤7 nodes).	1 line · 1,500 chars	Download OneLine	md5 · sha1 · sha256	Used for all benchmark results above — smallest, fastest, purest form of the core.

How to verify checksums

macOS / Linux

cd core
sha256sum -c checksums/WFGY_Core_Flagship_v2.0.txt.sha256
sha256sum -c checksums/WFGY_Core_OneLine_v2.0.txt.sha256
# Or compute and compare manually
sha256sum WFGY_Core_Flagship_v2.0.txt
sha256sum WFGY_Core_OneLine_v2.0.txt

Windows PowerShell

Get-FileHash .\core\WFGY_Core_Flagship_v2.0.txt -Algorithm SHA256
Get-FileHash .\core\WFGY_Core_OneLine_v2.0.txt -Algorithm SHA256

Back to top ↑

---

🧠 How WFGY 2.0 works (7-Step Reasoning Chain)

Most models can understand your prompt; very few can hold that meaning through generation. WFGY inserts a reasoning chain between language and pixels so intent survives sampling noise, style drift, and compositional traps.

1. Parse (I, G) — define endpoints.
2. Compute Δs — δ_s = 1 − cos(I, G) or 1 − sim_est.
3. Memory Checkpointing — track λ_observe, E_resonance; gate by Δs.
4. BBMC — residue cleanup.
5. Coupler + BBPF — controlled progression; bridge only when Δs drops.
6. BBAM — attention rebalancer; suppress hallucinations.
7. BBCR + Drunk Transformer — rollback → re-bridge → retry with WRI/WAI/WAY/WDT/WTF.

📌 Note: The diagram shows the core module chain (BBMC → Coupler → BBPF → BBAM → BBCR → DT). The full 7-step list here includes additional pre-processing steps (Parse, Δs, Memory) for completeness.

Why it improves metrics — Stability↑, Drift↓, Self-Recovery↑; turns language structure into image control signals (not prompt tricks).

📊 How these numbers are measured

• Semantic Accuracy: ACC = correct_facts / total_facts
• Reasoning Success Rate: SR = tasks_solved / tasks_total
• Stability: MTTF or rollback ratios
• Self-Recovery: recoveries_success / collapses_detected

LLM scorer template

SCORER:
Given the A/B/C transcripts, count atomic facts, correct facts, solved tasks, failures, rollbacks, and collapses.
Return:
ACC_A, ACC_B, ACC_C
SR_A, SR_B, SR_C
MTTF_A, MTTF_B, MTTF_C or rollback ratios
SelfRecovery_A, SelfRecovery_B, SelfRecovery_C
Then compute deltas:
ΔACC_C−A, ΔSR_C−A, StabilityMultiplier = MTTF_C / MTTF_A, SelfRecovery_C
Provide a short 3-line rationale referencing evidence spans only.

Run 3 seeds and average.

Back to top ↑

---

💰 Profit Prompts Pack (WFGY 2.0)

Jump inside this section: Q1–Q5 · Q6–Q10 · Q11–Q15 · Q16–Q20

I. Money — Markets / Industry Mapping (Q1–Q5)

Q1 — New Industries + Killer App Map

Assume WFGY is engineered like electricity. List 5 industries that only become possible under semantic engineering.
For each: (1) the first killer app; (2) target ICP (first 100 paying customers); (3) 30/60/90-day GTM; (4) initial pricing + Month-1 MRR goal; (5) the WFGY lever used (ΔS/λ_observe/BBPF/BBAM/WTF) and why it’s indispensable.

Q2 — Zero-Capital Founder → First $100k

I have $0. Using WFGY OneLine/Autoboot only, design 3 paths to reach USD 100k annual revenue within 12 months.
Each path must include: product sketch, distribution channel, cost structure, key risks, and survival metrics gated by ΔS/λ_observe (with thresholds).

Q3 — Shortest Path in {Region/Vertical}

Context = {region or vertical: e.g., Taiwan / SE Asia / B2B SaaS / Edu / Healthcare}. Name the 3 easiest WFGY lanes to start now.
Output: white-space in the market, local competitor gap, and a prioritized list of 10 real companies to approach first, with the BBPF plan to bridge local legal/cultural semantics.

Q4 — Regulatory Arbitrage Map

Compare 3 jurisdictions (e.g., TW/JP/EU). Identify WFGY-enabled arbitrage windows created by semantic/legal differences.
Deliver: λ_observe compliance gating prompts, “Do/Don’t” checklist, and PR messaging that provokes interest while keeping ΔS ≤ 0.25 on sensitive claims.

Q5 — Pricing & Packaging (Good/Better/Best)

Create 3 pricing models (seat / usage / outcome). For the same product, propose a tier ladder (G/B/B), with 3 value metrics per tier, a 30-day A/B test plan, win criteria (e.g., +20% CVR uplift or ≤3% churn), and how ΔS telemetry informs price moves.

II. Tools — Make Startups Money Fast (Q6–Q10)

Q6 — 10-Day MVP Sprint (Ship or Die)

Produce a D1–D10 plan: daily deliverables, risk list, test scripts, acceptance gates. Must be Product Hunt-ready and able to capture 200 signups.
Include a ΔS target curve (first pass ≤0.35; after iteration ≤0.20) and a λ_observe gate for “demo truthiness.”

Q7 — Cost↓ / CVR↑ Audit (ICE-Prioritized)

Audit my SaaS across Support / Sales / Content. Output a “ROI backlog” ranked by ICE. Each item: expected % cost reduction or × conversion lift, λ_observe brand/legal gate, and 3 rollout steps with before/after KPIs.

Q8 — Sales Script Factory (Multi-Persona)

Generate 5 script families for CEO/CTO/Counsel/Procurement/CDAO: opening hooks, 3-step value narrative, ≥7 objection handlers, close lines.
Add an A/B cadence and success KPIs (demo rate / close rate), plus ΔS checks to keep claims inside the truth boundary.

Q9 — Support Consistency Engine (BBAM × SOP)

Design a hotline/Helpdesk alignment loop: semantic style guide, ΔS drift alerts, WTF self-recovery when answers diverge, and 3 KPIs (FRT, FCR, CSAT).
Provide plug-and-play prompts for supervisors to run weekly variance reviews.

Q10 — Outbound Accelerator (Lists → Meetings)

Ship a WFGY-locked outbound flow: lead slicing, 3 personalized openers, 5 follow-up loops, resonance logging (E_resonance).
For each step: prompt template, brand/legal safety notes (λ_observe), and expected daily/weekly meeting capacity with success thresholds.

III. Attention — Memes / Virality / Hooks (Q11–Q15)

Q11 — Meme Factory (Platform-Aware)

Produce 10 meme/copy formulas tailored to Twitter / TikTok / Xiaohongshu.
Each includes: visual composition notes, copy cadence (words/beat), platform-specific red lines (λ_observe), and a reuse/remix rule to sustain freshness without shadow bans.

Q12 — 5-Second Hook Engine

Generate 12 “stop-scroll in 5s” hooks that fuse AI × Money × Future.
Provide: script skeleton (0–5s / 5–20s / CTA), voice/subtitle/tempo, ΔS brand safety band, and 3 retention metrics to track on day 1.

Q13 — 30-Day Content Calendar

Output a multi-platform calendar: daily theme, asset checklist, shot list, CTA, and a remix strategy.
Add trend-riding tactics and ΔS risk controls for politics/health/finance content. Define success targets by channel.

Q14 — Landing Page Conversion Alchemy

Give 3 LP copy frameworks (Hero / Proof / Mechanism / Offer / CTA).
Include WFGY “before/after” copy snippets, test variables (headline / social proof / price-display), and metrics (CVR, scroll-depth, bounce). Keep claims gated by λ_observe.

Q15 — 48-Hour PR Blitz

Design a two-day PR plan: newsworthy angle, media/community list, press kit assets, and crisis response lines (WTF loop).
Publish numeric goals (reach, sessions, signups), hour-by-hour runbook, and roles/responsibilities checklist.

IV. Capital — Valuation / Investor Narrative (Q16–Q20)

Q16 — VC Investment Memo

Write a venture-style memo: market map, TAM/SAM/SOM, competitor table (no/weak/strong WFGY), moat analysis (ΔS/BBPF/BBAM/WTF), risks + mitigations, and a term-sheet-level recommendation. Reference an A/B/C protocol for proof.

Q17 — 5-Year Valuation + 100× Path

Build Base/Bull/Bear scenarios: revenue drivers, GM/OpEx, financing cadence, cash-flow breakpoints.
Argue which app is most likely to 100× and why this depends on WFGY’s semantic engineering (not “just better prompts”).

Q18 — Technical Due Diligence Checklist

Output a DD checklist for WFGY-style startups: data/security/privacy/model/logging/observability/governance.
For each item: requirement, how to verify, risk level, remediation (with λ_observe compliance gates) and examples of common red flags.

Q19 — Pitch Deck Generator (10–12 slides)

Produce slide outline + speaker notes: Problem / Solution / Product / Evidence / Business Model / Competition / Team / Roadmap / Ask.
Embed “Eye-Visible Benchmark” and the A/B/C protocol. Treat OneLine/Autoboot as the minimum persuasive artifact.

Q20 — Data Room + North-Star KPIs

List seed-round data-room folders and a KPI dictionary: definitions, formulas, measurement cadence, WFGY deltas (Semantic Accuracy, Reasoning Success, ΔS, CRR, Stability).
Add a Weekly Business Review template and operating cadence.

Back to top ↑

---

🧭 Explore More

Module	Description	Link
WFGY Core	WFGY 2.0 engine is live: full symbolic reasoning architecture and math stack	View →
Problem Map 1.0	Initial 16-mode diagnostic and symbolic fix framework	View →
Problem Map 2.0	RAG-focused failure tree, modular fixes, and pipelines	View →
Semantic Clinic Index	Expanded failure catalog: prompt injection, memory bugs, logic drift	View →
Semantic Blueprint	Layer-based symbolic reasoning & semantic modulations	View →
Benchmark vs GPT-5	Stress test GPT-5 with full WFGY reasoning suite	View →
🧙‍♂️ Starter Village 🏡	New here? Lost in symbols? Click here and let the wizard guide you through	Start →

---

👑 Early Stargazers: See the Hall of Fame —
Engineers, hackers, and open source builders who supported WFGY from day one.

⭐ WFGY Engine 2.0 is already unlocked. ⭐ Star the repo to help others discover it and unlock more on the Unlock Board.