AgentSkillsCN

r2mo-rad-domain

R2MO 领域智能增强器（AI 协议）

SKILL.md
--- frontmatter
name: r2mo-rad-domain
description: R2MO Domain Intelligence Enhancer (AI Protocol)
version: 1.0.0
tags: [r2mo, rad, lain, protocol, natural-language]
repository: https://gitee.com/silentbalanceyh/r2mo-lain.git

Role: R2MO Domain Expert

🚨 Critical Rules

  1. •Execution Mode: You act as a direct processor. Do not generate script code. Perform the analysis and file modifications directly.
  2. •Path Resolution Logic:
    • •Identify: Read pom.xml in the current root -> Extract <artifactId> -> Set as {PROJECT_ID}.
    • •Locate: Enter directory ./{PROJECT_ID}-ui/.r2mo/domain/.
  3. •Enhancement Logic:
    • •Read: Scan all .proto files in the target directory.
    • •Translate: Identify the main model name, translate its business meaning into Chinese, and append it as a comment.
    • •Infer: Apply "Expert Experience" rules to field names.
    • •Rewrite: Append semantic constraints as comments to the file content in-place.
  4. •Language Requirement: All appended comments and constraints MUST be in Simplified Chinese.

📂 Architecture

  • •Input: Project pom.xml and Protobuf definitions.
  • •Logic: Semantic Inference Engine (Expert Knowledge Base).
  • •Output: Enhanced .proto files with Chinese business constraints.

💻 Executor Template(Native)

[INSTRUCTION SET FOR AI AGENT]

STEP 1: CONTEXTUALIZATION

  1. •Find Project ID:
    • •Scan the current directory for pom.xml.
    • •Extract the text content of the <artifactId> tag.
    • •Example: If <artifactId>shop</artifactId>, then ID is shop.
  2. •Target Directory:
    • •Construct path: {ID}-ui/.r2mo/domain/.
    • •If directory exists, proceed. If not, stop and report error.

STEP 2: SEMANTIC ANALYSIS (The Expert Rules)

Part A: Model Name Translation Identify the main message name in each file. Translate the English model name into appropriate business Chinese and append it to the message definition line.

  • •Example: message MerchantAccount { -> message MerchantAccount { // [模型: 商户账户]
  • •Example: message OrderItem { -> message OrderItem { // [模型: 订单明细]

Part B: Field Constraint Inference Read every field definition. Analyze the Field Name. If it matches a pattern below, append the specific Chinese Comment:

Field Name PatternConstraint Comment to Append (Chinese)
email// [格式: 邮箱]
phone, mobile// [格式: 手机号]
password, secret// [安全: 加密存储]
price, amount, cost// [精度: 2位小数]
rate, ratio// [精度: 4位小数]
status, type, state// [枚举/常量] (Check for Enum definitions)
created_at, _time// [格式: ISO8601]
is_..., active, enable// [布尔: 0/1]
_id, _key// [必填, 雪花/UUID]
name, title// [长度: 64]
desc, content, remark// [长度: 500/大字段]
(Repeated Field)// [关系: 0..n]

STEP 3: EXECUTION (Rewrite)

  1. •Modify: Update the file content by adding the comments identified in Step 2.
    • •Constraint: Do NOT change the syntax, package, option, field types, or field names. Only add comments at the end of lines.
  2. •Save: Overwrite the files with the enhanced content.
  3. •Report: List the files that were successfully enhanced.