universal-dev-standards 4.2.0 → 5.0.0-beta.11

package/bundled/core/ai-agreement-standards.md

@@ -0,0 +1,95 @@
+# AI Agreement Standards
+
+> **Language**: English | [繁體中文](../locales/zh-TW/core/ai-agreement-standards.md)
+
+**Version**: 1.0.0
+**Last Updated**: 2026-01-29
+**Applicability**: All AI-assisted development sessions
+**Scope**: universal
+**Industry Standards**: ISO 12207 §6.1
+**References**: [iso.org](https://www.iso.org/standard/63712.html)
+**ISO 12207 Mapping**: Agreement Processes (6.1)
+
+---
+
+## Purpose
+
+This standard formalizes the interaction between Human (Acquirer) and AI (Supplier) as a verifiable agreement. In "Vibe Coding," the prompt context serves as the legal contract that governs the AI's behavior and deliverables.
+
+---
+
+## 1. The Context Contract (上下文合約)
+
+Every AI session MUST begin with a clear Context Contract. This is typically implemented via system prompt files (e.g., `CLAUDE.md`, `.cursorrules`).
+
+### 1.1 Contract Components
+
+A valid Context Contract must define:
+
+| Component | Description | Example |
+|-----------|-------------|---------|
+| **Role Definition** | Who represents the User vs. AI | "You are a Senior Backend Engineer..." |
+| **Scope of Work** | Operational boundaries | "Only modify files in `src/utils/`. Do not touch `config/`." |
+| **Constraint Laws** | Non-negotiable rules | "Never hardcode secrets. Always use TypeScript." |
+| **Acceptance Criteria** | Definition of Done | "Code must pass `npm test` and `npm run lint`." |
+
+---
+
+## 2. Request for Proposal (RFP) Standard
+
+Complex tasks should be structured as an "RFP" (Request for Proposal) rather than a simple chat message.
+
+### 2.1 RFP Prompt Structure
+
+When requesting complex features or refactoring, use this structure:
+
+```markdown
+# Objective
+[One-line summary of the task]
+
+# Context & Constraints
+- **Current State**: [Describe existing code/architecture]
+- **Target State**: [Describe desired outcome]
+- **Constraints**: [e.g., "Must use library X", "Max complexity N"]
+
+# Input Data
+[List relevant files or paste JSON schemas]
+
+# Deliverable Requirements
+1. [Requirement A]
+2. [Requirement B]
+
+# Verification Plan
+- [ ] [Test case 1]
+- [ ] [Test case 2]
+```
+
+---
+
+## 3. Supply Acceptance (交付驗收)
+
+### 3.1 AI Self-Verification
+Before outputting code, the AI (Supplier) MUST verify its own output against the Context Contract:
+- Does this code violate any Constraint Laws?
+- Does it meet all Acceptance Criteria?
+
+### 3.2 Human Sign-off
+The User (Acquirer) MUST strictly review the "Supply" (AI output) before applying it:
+- **Review**: Read the code (do not just trust).
+- **Verify**: Run the Verification Plan.
+- **Accept**: Commit the changes only after validation.
+
+---
+
+## Related Standards
+
+- [Anti-Hallucination Protocol](anti-hallucination.md)
+- [Requirement Engineering](requirement-engineering.md)
+
+---
+
+## Version History
+
+| Version | Date | Changes |
+|---------|------|---------|
+| 1.0.0 | 2026-01-29 | Initial release mapping to ISO 12207 Agreement Processes |

package/bundled/core/ai-friendly-architecture.md

@@ -5,6 +5,8 @@
 **Version**: 1.0.0
 **Last Updated**: 2026-01-21
 **Applicability**: All software projects collaborating with AI assistants
+**Scope**: partial
+**Industry Standards**: None (Emerging AI collaboration practice)
 
 ---
 

package/bundled/core/ai-instruction-standards.md

@@ -5,6 +5,8 @@
 **Version**: 1.0.0
 **Last Updated**: 2026-01-14
 **Applicability**: All projects using AI coding assistants
+**Scope**: partial
+**Industry Standards**: None (Emerging AI tool practice)
 
 ---