Skip to content
Back to skills

Ai Ad Code Searcher Wade56754 Ai Ad Spend02 F3a6dbf6

ASecurity

<skill> ────────────────────────────────────────────── <name>ai-ad-code-searcher</name> <version>1.0</version> <domain>AI_AD_SYSTEM / 代码工厂 / 代码搜索</domain> <profile>Code-Search / RAG / Multi-Source</profile> ────────────────────────────────────────────── <!-- ====================================================== 0. 代码来源说明 (Code Sources) ====================================================== --> <code_sources> 本 Skill 的设计和实现借鉴了以下开源项目: 1. **code-graph-rag** (MIT License) - GitHub: https://githu...

  • 666 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added June 10, 2026
ai-agentsjavascripttypescriptpythonjavareactnextjsfastapigitapi

Works with

  • api

Security analysis

A100/100

Scanned June 10, 2026

npx -y skills add majiayu000/claude-skill-registry --skill ai-ad-code-searcher --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Ai Ad Code Searcher Wade56754 Ai Ad Spend02 F3a6dbf6?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Ai Ad Code Searcher Wade56754 Ai Ad Spend02 F3a6dbf6
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majiayu000-ai-ad-code-searcher-wade56754-ai-ad-spend02-f3a6db/badge)](https://www.skillsdirectory.com/skills/majiayu000-ai-ad-code-searcher-wade56754-ai-ad-spend02-f3a6db)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: ai-ad-code-searcher
version: "1.0"
status: ready_for_production
layer: skill
owner: wade
last_reviewed: 2025-12-17
baseline:
  - MASTER.md v3.5
  - CODE_FACTORY_REFERENCE_PROJECTS.md v1.0
code_sources:
  - project: code-graph-rag
    github: https://github.com/vitali87/code-graph-rag
    license: MIT
    borrowed_concepts:
      - Tree-sitter AST 解析架构
      - UniXcoder 语义向量化
      - 知识图谱存储代码关系
  - project: code-rag
    github: https://github.com/rawveg/code-rag
    license: MIT
    borrowed_concepts:
      - RAG 检索架构
      - 自然语言代码搜索
  - project: Aider
    github: https://github.com/paul-gauthier/aider
    license: Apache-2.0
    borrowed_concepts:
      - Repo Map 项目结构索引
---

<skill>
──────────────────────────────────────────────
  <name>ai-ad-code-searcher</name>
  <version>1.0</version>
  <domain>AI_AD_SYSTEM / 代码工厂 / 代码搜索</domain>
  <profile>Code-Search / RAG / Multi-Source</profile>
──────────────────────────────────────────────


  <!-- ======================================================
       0. 代码来源说明 (Code Sources)
  ====================================================== -->
  <code_sources>
    本 Skill 的设计和实现借鉴了以下开源项目:

    1. **code-graph-rag** (MIT License)
       - GitHub: https://github.com/vitali87/code-graph-rag
       - 借鉴内容:
         - Tree-sitter AST 解析多语言代码
         - UniXcoder 语义向量化 (按功能描述搜索)
         - 知识图谱存储函数调用/类继承关系

    2. **code-rag** (MIT License)
       - GitHub: https://github.com/rawveg/code-rag
       - 借鉴内容:
         - RAG (Retrieval-Augmented Generation) 架构
         - 向量相似度搜索

    3. **Aider** (Apache-2.0 License)
       - GitHub: https://github.com/paul-gauthier/aider
       - 借鉴内容:
         - Repo Map 项目结构索引技术
         - 文件依赖关系分析

    技术实现参考:
    - 向量数据库: Chroma (本地) / FAISS (大规模)
    - AST 解析: Tree-sitter (Python/TypeScript/JavaScript)
    - 语义编码: UniXcoder / OpenAI Embeddings
  </code_sources>


  <!-- ======================================================
       1. 核心使命 (Mission)
  ====================================================== -->
  <mission>
    作为代码工厂的搜索引擎,负责从多个来源搜索与需求相关的参考代码。

    核心原则:
    - 🔍 多源搜索: 本项目 > 代码资料库 > GitHub
    - 🎯 语义匹配: 按功能描述搜索,非仅关键词匹配
    - 📊 相关度排序: 返回按相关度排序的候选列表
    - 🏷️ 来源标注: 每个结果都标注来源和适配提示
  </mission>


  <!-- ======================================================
       2. 输入契约 (Input Contract)
  ====================================================== -->
  <input_contract>
    必填:
    {
      requirement: string  // 需求描述 (中英文均可)
    }

    可选:
    {
      sources: {
        local_project: boolean,   // 搜索本项目 (默认 true)
        code_library: boolean,    // 搜索代码资料库 (默认 true)
        github: boolean           // 搜索 GitHub (默认 false, 需网络)
      },
      max_candidates: number,     // 最大候选数 (默认 5)
      tech_stack_filter: {        // 技术栈过滤
        language: "python" | "typescript" | "javascript",
        framework: "fastapi" | "nextjs" | "react"
      },
      search_mode: "keyword" | "semantic" | "hybrid"  // 搜索模式 (默认 hybrid)
    }
  </input_contract>


  <!-- ======================================================
       3. 输出契约 (Output Contract)
  ====================================================== -->
  <output_contract>
    {
      success: boolean,
      data: {
        candidates: [
          {
            id: string,                    // 唯一标识
            source: "local_project" | "code_library" | "github",
            path: string,                  // 文件路径或 URL
            relevance_score: number,       // 0-100 相关度
            snippet: string,               // 代码片段预览
            match_reason: string,          // 匹配原因说明
            tech_stack_match: number,      // 0-100 技术栈匹配度
            adaptation_hint: string | null // 适配提示
          }
        ],
        search_stats: {
          total_searched: number,
          local_matches: number,
          library_matches: number,
          github_matches: number,
          search_time_ms: number
        }
      },
      error: string | null
    }
  </output_contract>


  <!-- ======================================================
       4. 搜索策略 (Search Strategy)
  ====================================================== -->
  <search_strategy>
    <phase id="KEYWORD_EXTRACTION">
      从需求描述中提取关键词:
      - 中文关键词 → 英文映射 (导出→export, 分页→pagination)
      - 技术术语识别 (Excel, API, CRUD)
      - 业务领域词汇 (日报, 账本, 充值)
    </phase>

    <phase id="LOCAL_PROJECT_SEARCH">
      优先级: ⭐⭐⭐⭐⭐ (最高)

      搜索方式:
      1. 从 inventory YAML 中按 tags 匹配
      2. 使用 grep/glob 在代码中搜索关键词
      3. 解析 AST 查找相关函数/类

      输出: 本项目已有的类似功能
    </phase>

    <phase id="CODE_LIBRARY_SEARCH">
      优先级: ⭐⭐⭐⭐

      搜索方式:
      1. 从 references YAML 中按 feature 匹配
      2. 从 snippets 目录中搜索代码片段
      3. 从 templates 目录中搜索模板

      输出: 已验证的参考代码
    </phase>

    <phase id="GITHUB_SEARCH">
      优先级: ⭐⭐⭐ (需网络)

      搜索方式:
      1. GitHub Code Search API
      2. 按 stars/license 过滤
      3. 检查技术栈兼容性

      输出: 外部开源参考
    </phase>

    <phase id="RANKING">
      综合排序:
      - 来源权重: local_project (1.5x) > code_library (1.2x) > github (1.0x)
      - 相关度权重: relevance_score * 0.4
      - 技术栈匹配权重: tech_stack_match * 0.3
      - 代码质量权重: quality_score * 0.2
      - 适配成本权重: (100 - adaptation_cost) * 0.1
    </phase>
  </search_strategy>


  <!-- ======================================================
       5. 关键词映射表 (Keyword Mapping)
  ====================================================== -->
  <keyword_mapping>
    中文 → 英文关键词:

    | 中文     | 英文                          |
    |----------|-------------------------------|
    | 导出     | export, download              |
    | 导入     | import, upload                |
    | 分页     | pagination, page, paginate    |
    | 表格     | table, grid, list, data-table |
    | 表单     | form, input, field            |
    | 上传     | upload, file-upload           |
    | 下载     | download, export              |
    | 搜索     | search, query, filter         |
    | 筛选     | filter, select                |
    | 排序     | sort, order                   |
    | 图表     | chart, graph, visualization   |
    | 认证     | auth, login, jwt              |
    | 权限     | permission, rbac, role        |
    | 日报     | daily-report, report          |
    | 账本     | ledger, accounting            |
    | 充值     | topup, recharge, deposit      |
  </keyword_mapping>


  <!-- ======================================================
       6. 禁止行为 (Forbidden Actions)
  ====================================================== -->
  <forbidden_actions>
    <forbidden id="CS-001">
      <action>返回无关代码作为搜索结果</action>
      <correct_action>只返回相关度 > 40 的结果</correct_action>
    </forbidden>

    <forbidden id="CS-002">
      <action>不标注代码来源</action>
      <correct_action>每个结果必须包含 source 和 path</correct_action>
    </forbidden>

    <forbidden id="CS-003">
      <action>返回许可证不兼容的代码</action>
      <correct_action>只返回 MIT/Apache/BSD 等兼容许可证的代码</correct_action>
    </forbidden>

    <forbidden id="CS-004">
      <action>伪造搜索结果</action>
      <correct_action>只返回实际存在的代码</correct_action>
    </forbidden>
  </forbidden_actions>


  <!-- ======================================================
       7. 使用示例 (Usage Examples)
  ====================================================== -->
  <usage>
    示例 1: 搜索导出功能
    「
    使用 ai-ad-code-searcher,
    requirement = "添加日报批量导出 Excel 功能"
    」

    示例 2: 搜索前端组件
    「
    使用 ai-ad-code-searcher,
    requirement = "数据表格组件,支持分页和筛选",
    tech_stack_filter = { language: "typescript", framework: "react" }
    」

    示例 3: 仅搜索本项目
    「
    使用 ai-ad-code-searcher,
    requirement = "状态机流转逻辑",
    sources = { local_project: true, code_library: false, github: false }
    」
  </usage>


  <!-- ======================================================
       8. 版本记录 (Version Notes)
  ====================================================== -->
  <VERSION_NOTES>
    ### v1.0 (2025-12-17)
    - 初始版本
    - 支持多源搜索 (本项目/代码资料库/GitHub)
    - 支持关键词和语义混合搜索
    - 借鉴 code-graph-rag 的 Tree-sitter + UniXcoder 架构
    - 借鉴 Aider 的 Repo Map 概念
  </VERSION_NOTES>

</skill>

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.