T
Tianxia-Core/Build-Law
Build Law v1.2.0 — 4-Pillar, 14-Gate Deterministic AI Engineering Constitution, 5.0x–8.8x Token-Efficiency…
更新 2026-11-19
在 GitHub 查看 ↗直链 · 无中转页
- ★ 数取值
- 0 ★ · Tianxia-Core/Build-Law · 取自 GitHub 搜索 API · (2026-11-19 20:00:48 核对)
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型:开源项目
- 收录与活跃
- 收录于 2026-11-19 · 最近见到 2026-11-20 · 共见到 2 次 · 被 1 个来源列出
- 评分构成
- 综合分 103 · 1 个来源 · repo
常见问题(答案都来自我们的记录)
它适合谁 / 用来做什么?
我们把它归在这些标签下: 开源项目。标签来自条目自身,不是我们强加的分类。
谁维护它、还在更新吗?
仓库
Tianxia-Core/Build-Law;我们收录于 2026-11-19;最近见到 2026-11-20。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。README 原文(英文,未翻译) · 在 GitHub 读完整 README →
# Build Law
以下摘录自该仓库的 README(片段)
The Constitutional Engineering Standard, 14-Gate Static Verifier & 5.0×–8.8× Token-Efficiency Framework for AI-Native Software Development
Drop-in support for: Claude Code · Claude.ai Skills · Cursor · OpenCode · VS Code (.vsix) · GitHub Copilot · Windsurf · Cline · Roo Code · Continue · Aider · Zed · JetBrains Junie · Gemini CLI · OpenAI Codex · Goose · MCP (stdio) · Nix Flakes · GitHub Actions
Plain-English Guide (Layman Section) ·
60-Second Install ·
Engineering Philosophy Showcase ·
Measured Benchmarks ·
Work With Us / Hire / Partner
## 🌍 In Plain English — What Build-Law Is & Why It Matters (Layman's Guide)
The Everyday Problem: AI coding assistants frequently leave unfinished TODO placeholders, accidentally expose API keys, bloat files until context windows collapse, and burn 5×–8× more API tokens than necessary.
How It Works (Without the Jargon): Build Law gives every AI model and IDE a clear 4-Pillar, 14-Gate engineering constitution backed by a
## TL;DR (What Is Build Law?)
Build Law is an open-source (MIT) constitutional engineering system and zero-token static verifier created by Collins Takem at RASA-X-MACHINA to govern how tianxiaOS and its polyglot autonomous organisms are built.
Instead of treating AI coding assistants like slot machines wrapped in vague "act as a senior engineer" prompts, Build Law binds every model, IDE, CLI agent, and CI pipeline to four hard engineering laws:
1. 14-Gate Completeness (DSS-14.2): Zero TODO stubs, zero eval()/shell injection, bounded loop timeouts, and explicit error recovery before a file is allowed to ship.
2. Polyglot FFP-First Modularity ( S2["2. Mine
(Rank = P+C+A+E+F)"]
S2 --> S3["3. Decompose
( S4["4. Gate
(14-Lens Check)"]
S4 --> S5["5. Package
(Ed25519 Signed Formats)"]
S5 --> S6["6. Deploy
(CLI · Web · TUI · GUI)"]
S6 --> S7["7. Attest
(Merkle Audit Seal)"]
S7 -.-> S1
## 📊 Measured Production Telemetry (Before vs. After Build Law)
Every claim below is measured directly against live production subsystems in the @Tianxia-Core ecosystem:
| Engineering Dimension | Ungoverned / Pre-Ratchet Baseline | Under Build Law (v5.1) | Verified Production Impact |
| Structured CRM / Data Query (2,565 Entities) | Raw corpus prompt scan (743,321 tokens ≈ $2.23/call) | Vāk Rust AST → SQLite FTS5 (0 LLM tokens; 40–107 µs parse) | 100% token elimination ($0.00) |
Drop-in support for: Claude Code · Claude.ai Skills · Cursor · OpenCode · VS Code (.vsix) · GitHub Copilot · Windsurf · Cline · Roo Code · Continue · Aider · Zed · JetBrains Junie · Gemini CLI · OpenAI Codex · Goose · MCP (stdio) · Nix Flakes · GitHub Actions
Plain-English Guide (Layman Section) ·
60-Second Install ·
Engineering Philosophy Showcase ·
Measured Benchmarks ·
Work With Us / Hire / Partner
## 🌍 In Plain English — What Build-Law Is & Why It Matters (Layman's Guide)
The Everyday Problem: AI coding assistants frequently leave unfinished TODO placeholders, accidentally expose API keys, bloat files until context windows collapse, and burn 5×–8× more API tokens than necessary.
How It Works (Without the Jargon): Build Law gives every AI model and IDE a clear 4-Pillar, 14-Gate engineering constitution backed by a
## TL;DR (What Is Build Law?)
Build Law is an open-source (MIT) constitutional engineering system and zero-token static verifier created by Collins Takem at RASA-X-MACHINA to govern how tianxiaOS and its polyglot autonomous organisms are built.
Instead of treating AI coding assistants like slot machines wrapped in vague "act as a senior engineer" prompts, Build Law binds every model, IDE, CLI agent, and CI pipeline to four hard engineering laws:
1. 14-Gate Completeness (DSS-14.2): Zero TODO stubs, zero eval()/shell injection, bounded loop timeouts, and explicit error recovery before a file is allowed to ship.
2. Polyglot FFP-First Modularity ( S2["2. Mine
(Rank = P+C+A+E+F)"]
S2 --> S3["3. Decompose
( S4["4. Gate
(14-Lens Check)"]
S4 --> S5["5. Package
(Ed25519 Signed Formats)"]
S5 --> S6["6. Deploy
(CLI · Web · TUI · GUI)"]
S6 --> S7["7. Attest
(Merkle Audit Seal)"]
S7 -.-> S1
## 📊 Measured Production Telemetry (Before vs. After Build Law)
Every claim below is measured directly against live production subsystems in the @Tianxia-Core ecosystem:
| Engineering Dimension | Ungoverned / Pre-Ratchet Baseline | Under Build Law (v5.1) | Verified Production Impact |
| Structured CRM / Data Query (2,565 Entities) | Raw corpus prompt scan (743,321 tokens ≈ $2.23/call) | Vāk Rust AST → SQLite FTS5 (0 LLM tokens; 40–107 µs parse) | 100% token elimination ($0.00) |
收录徽章(可选)
想把收录标记放进 README 就复制这行,可选,不影响收录。
[](https://anclum.com/zh/s/tianxia-core-build-law/)
- 来源
- github · https://github.com/Tianxia-Core/Build-Law
- 按领域浏览
- 开源项目
- 收录于
- 今天
更多条目
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。