CZ (zccz14) Personal Knowledge Base Objective Analysis Report

AI Analysis Time: February 8, 2026 Generated from 68 Markdown files Note: This report is AI-generated and its content is for reference only.

Overview

Repository owner CZ (GitHub ID: zccz14) describes their professional domain as quantitative trading and fintech. They are the founder of the open-source project No Trade No Life (NTNL) and hold a bachelor's degree from Xi'an Jiaotong University (XJTU). Their native language is Chinese, they can communicate in English and Japanese, and are currently learning Spanish.

The repository content spans from January 5, 2026, to February 8, 2026, approximately 35 days. The content is written in Chinese and distributed across the following directories:

• LOGS (54 entries): Work logs documenting daily development, reflections, and team collaboration.
• INSIGHTS (9 entries): Refined, in-depth articles covering investment theory, AI system design, cognitive philosophy, etc.
• QUANT (2 entries): Technical documents related to quantitative trading (FSA Full Spectrum Analysis, EA project introduction).
• MEETINGS (1 entry): Meeting minutes (analysis of Guotai Haitong private fund).
• EVENTS (1 entry): Event analysis (U.S. government seizure of Bitcoin case).
• DEBATES (1 entry): AI-generated debate text (Steady Growth vs. Class Leap).

Major projects include: CZON/CZONE (personal knowledge base tool), EA/Earnby.AI (on-chain fund project), Yuan/1earn (quantitative trading platform), LegionMind (AI Agent project management tool), FSA (Full Spectrum Analysis), SandTable (Capital Protracted War experimental framework).

Project Introduction

CZON / CZONE

• Origin: Started around January 7, 2026, initially named ZEN, later renamed to CZON (CZ + Zone) due to naming conflicts. Positioned as an AI-native personal knowledge base and content publishing tool.
• Core Features:
  • Write in native language using Markdown, with AI automatically translating into multilingual versions.
  • AI automatically extracts document metadata (title, summary, keywords).
  • AI automatic classification and tagging system.
  • AI generates multi-style summary reports (objective, critical, literary, personality analysis, etc.).
  • Dark mode, sitemap generation, SEO optimization.
  • Adversarial generation translation (translation Agent + review Agent iteration).
• Version Iteration: From 0.4.x to 0.8.6, underwent multiple iterations including JSX rendering refactoring, directory structure refactoring (from Hash ID to path mapping), YAML Front Matter removal, OpenCode Agent integration and rollback.
• CZONE: The online version of CZON (CZON + Environment). Plans to host via GitHub Pages, authorize via GitHub OAuth, run CLI via GitHub Actions, achieving a zero-server-cost online content creation platform.
• Related Files: From Creation to Distribution—Building an AI-Native Content Engine, CZON Meaning Interpretation, N-Shaped Potential Energy Curve

EA (Earnby.AI)

• Origin: Formerly a private fund operating from March 2024, launched as an on-chain project in November 2025, deployed on BSC (BNB Smart Chain).
• Core Features:
  • Stablecoin principal-protected wealth management + floating yield.
  • Priority/subordinated capital structure, with project's own capital fully invested in the subordinated tranche.
  • Directional portfolio strategy (machine learning timing) and Delta neutral strategy (cross-market arbitrage).
• Version Iteration: Cumulative return +39.22%, annualized ~22%, maximum drawdown 1.12%, profitable month ratio 95% (19/20 months).
• Related Files: EA Project Introduction

Yuan / 1earn

• Origin: Internal team quantitative trading infrastructure platform.
• Core Features:
  • Development, backtesting, and deployment of quantitative trading strategies.
  • Delta neutral strategy execution (spread trading, funding rate arbitrage, etc.).
  • 1earn is a user-facing product developed based on Yuan.
• Version Iteration: Continuous fixes for API rate limiting and stability issues during January 2026.
• Related Files: Team Work Log

LegionMind

• Origin: AI Agent project management tool led by team member C1.
• Core Features:
  • Cross-session context management based on file system.
  • Integrated GitHub Project Kanban functionality.
  • Supports scheduling and coordination of Agent tasks.
• Related Files: LegionMind Discussion, RFC Discussion

FSA (Full Spectrum Analysis)

• Origin: An extension based on J.L. Kelly's 1956 paper, a probabilistic strategy framework for investment trading scenarios.
• Core Features:
  • Defines outcome space, probability distribution, and return rate to calculate optimal leverage.
  • Uses Newton's method to solve optimization problems for strictly concave functions.
  • Includes black swan defense mechanism (symmetric pseudo-probability injection).
  • Gross Profit Margin (GPM) backtesting method and position resolution concept.
• Related Files: Full Spectrum Analysis: The Optimal Method for Monetizing Information

SandTable

• Origin: Named on February 4, 2026, an experimental framework for validating the "Capital Protracted War" theory.
• Core Features:
  • Market sequence generation (GBM, GARCH, Heston, etc. models).
  • Separation of signal strategy and betting strategy.
  • Evaluates profit-taking frequency and time intervals under different target multiples.
• Version Iteration: Published to npm (package name sandt).
• Related Files: Naming SandTable, Experimental Design

Key Themes

1. Investment Theory and Capital Management

• Capital Protracted War: Proposes a strategic framework for individual investors to achieve wealth leapfrogging through "controllable losses, cumulative advantage, and pressing the advantage." Refutes three viewpoints: "individuals are doomed to fail," "all-in for sudden wealth," and "steady development." Emphasizes programmatic trading and pyramiding.
• Full Spectrum Analysis: The Optimal Method for Monetizing Information: Extends the Kelly Criterion to multi-outcome scenarios, proposes optimal leverage calculation method, GPM backtesting framework, and black swan defense mechanism.
• Three-Body Dynamics Hypothesis of Capital Markets: Classifies market participants into three types: momentum capital (M), value capital (V), liquidity capital (L), constructing 8 market phases and phase transition models.
• Volatility and Leverage: Argues that leverage can be reduced to volatility, analyzes the nature and cost of on-exchange/off-exchange leverage.
• Volatility Effect is Superior to Leverage Effect: Argues that volatility provides traders with free spreads, while leverage linearly increases transaction costs.
• Capital Protracted War Experimental Design: Defines a separation framework for base account and betting account, formulates anti-Martingale position management rules.

2. AI System Design and Multi-Agent Collaboration

• Module-Level Human-AI Collaborative Software Engineering Architecture: Proposes an Agent workflow: Protocol Spec → parallel generation of implementation/test/benchmark code → multi-level arbitration.
• How to Solve Human Desire for Control—On the Controllable Trust Problem in Human-AI Collaboration: Proposes a two-layer multiplicative model for controllable trust (intent alignment × risk control triangle) and a fractal recursive structure for intent alignment.
• Embracing "Finite," Designing "Infinite"—A New Paradigm for Constructing Agent Systems Based on LLM Constraints: Proposes a trinity framework: Coordination Engineering, AI Decision Economics, Cognitive Flow Management.
• From Battlefield to Digital Space: On the Inspirational Significance of Su Yu's Combat Instructions for Multi-Agent System Collaboration Frameworks: Transfers the organizational logic of Su Yu's large-scale troop combat instructions to AI multi-agent collaboration frameworks.
• Multi-Agents Control Constraints: Discusses the trade-off between soft constraints (Agent controlling Agent) and hard constraints (Script controlling Agent).

3. AI Programming Practice and Engineering Reflection

• Vibe Coding Major Failure: Records the experience of AI-generated code quality collapse, suggests avoiding OOP and following Occam's razor principle.
• The Dawn of Liberation is Approaching: Discusses the user experience of vibe-kanban tool and the issue of desire for control.
• LLM Intelligence Level vs. Engineering Methods: Records discussion with Hobo on "nativism" vs. "constructivism," and the necessity of observability in production environments.
• Agent Translation Task Performance Inferior to One-Shot LLM: Records regression in Agent translation tasks and analysis of reasons.
• AI Programming Progress Too Fast, Foundation Unstable: Analyzes OpenCode's shortcomings in debugging ability, attention allocation, and architectural rhythm.

4. Cognitive Philosophy and Personal Existence

• On the Essence of Humanity: Explores personal knowledge base as a long-term memory carrier, the non-replicability of subjectivity, layered readers for "establishing words," the essence of taste (the ability to reject), and the deconstruction and reconstruction of personal meaning in the AI era.
• Returning to Simplicity: Complexity is an Inevitable Path of Cognition: Argues for the spiral ascent process of cognitive development, proposes "the tuition of complexity has a distance decay effect."
• From Creation to Distribution—Building an AI-Native Content Engine: Proposes the content creation principle of "deep in creation, shallow in distribution."

5. CZON Product Development

• ZEN Renamed to CZON: Records the naming process and product positioning thoughts.
• CZON's N-Shaped Potential Energy Curve: Proposes a potential energy change model from content creation to distribution.
• CZON Meaning Interpretation: Defines C (Content Oriented), Z (Zero Configuration), O (Organic AI-Native), N (N-shaped Energy Curve).
• CZON Directory Structure Refactoring: Refactoring from Hash ID to path mapping, solving the avalanche regeneration problem.
• CZON v0.8.6 Release: Removes YAML Front Matter, switches to JSON metadata translation.

6. Market Events and Industry Analysis

• How Did the U.S. Government Seize Chen Zhi's Approximately $15 Billion in Bitcoin?: Details the complete process of on-chain tracking, legal procedures, and private key recovery.
• In-Depth Analysis of Guotai Haitong 2025 Securities Private Funds: Covers private market landscape, strategy performance divergence, quantitative factor changes, allocation structure changes, etc.
• "Qi Pa Shuo" 4v4 Special Edition: Steady Growth vs. Class Leap: AI-generated debate text, revolving around the core ideas of Capital Protracted War.

Timeline

Key Individuals

Theoretical Framework Summary

Capital Protracted War

• Core Idea: Individual investors achieve class leapfrogging by "using controllable losses to bet on an unaffordable gain."
• Key Elements:
  • Four Principles: No risk (control loss rate), no waste (programmatic trading), no delay (pyramiding), no ambiguity (define victory conditions).
  • Risk Control Line: $F(t) \ge -C \cdot (t - T_0)$, where $C$ is the maximum bearable loss rate.
  • Pyramiding Strategy: Increase position size when profitable, return to risk control line and wait when losing.
  • Evaluation Metric: $T_S(A, C)$, the time required for a strategy to achieve the victory target given risk control parameters.
• Mathematical Expression: $F(t_1) = F(t_0) - P(t_0) + S(P(t_0), t_0, t_1)$
• Source File: Capital Protracted War

Full Spectrum Analysis (FSA)

• Core Idea: Calculate the optimal leverage that maximizes compound return rate within a given outcome space.
• Key Elements:
  • Outcome space $X$ satisfies five assumptions: deterministic return, deterministic probability, mutual exclusivity, completeness, leverage effect.
  • Compound Return Rate $C(X, k) = (\prod_{i \in X} (1 + k \cdot R_i)^{P_i}) - 1$
  • $G(k)$ is a strictly concave function, has a unique maximum point.
  • Black Swan Defense: Inject symmetric pseudo-probability events, restrict feasible domain to $(-1, 1)$.
• Mathematical Expression: $k_o = \argmax_k C(X, k)$, solved via Newton's method.
• Source File: Full Spectrum Analysis: The Optimal Method for Monetizing Information

Three-Body Dynamics Hypothesis of Capital Markets

• Core Idea: Capital markets are a three-body system composed of momentum capital (M), value capital (V), and liquidity capital (L), generating complex dynamic behaviors through interaction.
• Key Elements:
  • Three core variables: Premium $\delta$, Momentum $\mu$, Volatility $\sigma$.
  • Return-Risk-Cost matrix has an axisymmetric structure.
  • 8 market phases ($2^3$ combinations) and 12 single-dimension transitions.
  • Positive feedback loops (collapse spiral) and negative feedback loops (recovery mechanism).
  • Return-driven natural selection as a slow variable mechanism.
• Source File: Three-Body Dynamics Hypothesis of Capital Markets

Controllable Trust Two-Layer Multiplicative Model

• Core Idea: The desire for control in human-AI collaboration stems from rational concern about losing control of consequences. Achieve scalable collaboration by constructing controllable trust.
• Key Elements:
  • Base Layer: Intent Alignment (expression alignment × value alignment × structural alignment × dynamic alignment).
  • Execution Layer: Risk Control Triangle (predictability × intervenability × recoverability).
  • Fractal recursive structure of intent alignment.
  • Well-Organized Agent implementation framework.
• Mathematical Expression: Controllable Trust = Intent Alignment Index × Risk Control Index
• Source File: How to Solve Human Desire for Control—On the Controllable Trust Problem in Human-AI Collaboration

N-Shaped Potential Energy Curve

• Core Idea: The potential energy change from content creation to distribution follows an N-shaped polyline.
• Key Elements:
  • Low starting point (inspiration germination) → Rising potential energy (refined crafting) → Falling potential energy (dimensionality reduction distribution) → Rising potential energy again (audience resonance).
• Source File: N-Shaped Potential Energy Curve

Coordination Engineering / AI Decision Economics / Cognitive Flow Management

• Core Idea: The key to building powerful AI systems lies in accepting the "finiteness" of LLMs, transforming limitations into rules that drive evolution through system design.
• Key Elements:
  • Coordination Engineering: Checklist mode, parliamentary debate mode, constraint solver mode.
  • AI Decision Economics: Base currency layer, value assessment layer, decision strategy layer, market coordination layer.
  • Cognitive Flow Management: From "indoctrination" to "navigation," progressive cognitive loading, iterative alignment loops.
• Source File: Embracing "Finite," Designing "Infinite"—A New Paradigm for Constructing Agent Systems Based on LLM Constraints