@chongyan/autospec 1.0.1

package/knowledge/environment/detection-patterns.yaml

@@ -0,0 +1,502 @@
+# AutoSpec 检测规则配置
+# 用于技术栈、框架、AI组件的自动检测
+
+# ============================================================
+# 构建文件规则
+# ============================================================
+buildFiles:
+  package.json:
+    buildTool: npm/yarn/pnpm
+    lang: javascript
+    parser: packageJson
+
+  pom.xml:
+    buildTool: maven
+    lang: java
+    parser: pomXml
+
+  build.gradle:
+    buildTool: gradle
+    lang: java
+    parser: buildGradle
+
+  build.gradle.kts:
+    buildTool: gradle
+    lang: kotlin
+    parser: buildGradleKts
+
+  go.mod:
+    buildTool: go-modules
+    lang: go
+    parser: goMod
+
+  Cargo.toml:
+    buildTool: cargo
+    lang: rust
+    parser: cargoToml
+
+  pyproject.toml:
+    buildTool: poetry/pdm
+    lang: python
+    parser: pyprojectToml
+
+  requirements.txt:
+    buildTool: pip
+    lang: python
+    parser: requirements
+
+  setup.py:
+    buildTool: setuptools
+    lang: python
+    parser: setupPy
+
+  dbt_project.yml:
+    buildTool: dbt
+    lang: sql
+    parser: dbtProject
+
+  pubspec.yaml:
+    buildTool: flutter
+    lang: dart
+    parser: pubspec
+
+  composer.json:
+    buildTool: composer
+    lang: php
+    parser: composer
+
+  Gemfile:
+    buildTool: bundler
+    lang: ruby
+    parser: gemfile
+
+# ============================================================
+# 框架识别规则
+# ============================================================
+frameworks:
+  # JavaScript/TypeScript 框架
+  javascript:
+    - deps: [react-native]
+      framework: React Native
+      type: mobile
+
+    - deps: [next]
+      framework: Next.js
+      type: frontend
+
+    - deps: [nuxt]
+      framework: Nuxt.js
+      type: frontend
+
+    - deps: [vue]
+      framework: Vue
+      type: frontend
+
+    - deps: [react]
+      framework: React
+      type: frontend
+
+    - deps: [angular, "@angular/core"]
+      framework: Angular
+      type: frontend
+
+    - deps: [svelte]
+      framework: Svelte
+      type: frontend
+
+    - deps: [sveltekit]
+      framework: SvelteKit
+      type: frontend
+
+    - deps: [astro]
+      framework: Astro
+      type: frontend
+
+    - deps: [remix]
+      framework: Remix
+      type: frontend
+
+    - deps: [express]
+      framework: Express
+      type: backend
+
+    - deps: [nestjs, "@nestjs/core"]
+      framework: NestJS
+      type: backend
+
+    - deps: [fastify]
+      framework: Fastify
+      type: backend
+
+    - deps: [koa]
+      framework: Koa
+      type: backend
+
+    - deps: [hono]
+      framework: Hono
+      type: backend
+
+    - deps: [hapi]
+      framework: Hapi
+      type: backend
+
+    - deps: [openai]
+      framework: OpenAI
+      type: llm
+
+    - deps: [anthropic, "@anthropic-ai/sdk"]
+      framework: Anthropic
+      type: llm
+
+    - deps: [langchain, "@langchain/core"]
+      framework: LangChain
+      type: llm
+
+    - deps: [llamaindex]
+      framework: LlamaIndex
+      type: llm
+
+    - deps: [electron]
+      framework: Electron
+      type: desktop
+
+  # Python 框架
+  python:
+    - deps: [django]
+      framework: Django
+      type: backend
+
+    - deps: [fastapi]
+      framework: FastAPI
+      type: backend
+
+    - deps: [flask]
+      framework: Flask
+      type: backend
+
+    - deps: [starlette]
+      framework: Starlette
+      type: backend
+
+    - deps: [tornado]
+      framework: Tornado
+      type: backend
+
+    - deps: [sanic]
+      framework: Sanic
+      type: backend
+
+    - deps: [torch]
+      framework: PyTorch
+      type: ai
+
+    - deps: [tensorflow]
+      framework: TensorFlow
+      type: ai
+
+    - deps: [transformers]
+      framework: Transformers
+      type: ai
+
+    - deps: [langchain]
+      framework: LangChain
+      type: llm
+
+    - deps: [llamaindex]
+      framework: LlamaIndex
+      type: llm
+
+    - deps: [openai]
+      framework: OpenAI
+      type: llm
+
+    - deps: [anthropic]
+      framework: Anthropic
+      type: llm
+
+    - deps: [crewai]
+      framework: CrewAI
+      type: agent
+
+    - deps: [autogen]
+      framework: AutoGen
+      type: agent
+
+    - deps: [semantic-kernel]
+      framework: Semantic Kernel
+      type: agent
+
+    - deps: [dbt]
+      framework: DBT
+      type: data
+
+    - deps: [airflow]
+      framework: Airflow
+      type: data
+
+    - deps: [prefect]
+      framework: Prefect
+      type: data
+
+    - deps: [dagster]
+      framework: Dagster
+      type: data
+
+    - deps: [pyspark]
+      framework: PySpark
+      type: data
+
+    - deps: [dask]
+      framework: Dask
+      type: data
+
+    - deps: [faiss]
+      framework: FAISS
+      type: vector-store
+
+    - deps: [chromadb]
+      framework: ChromaDB
+      type: vector-store
+
+    - deps: [pinecone]
+      framework: Pinecone
+      type: vector-store
+
+    - deps: [weaviate]
+      framework: Weaviate
+      type: vector-store
+
+    - deps: [qdrant]
+      framework: Qdrant
+      type: vector-store
+
+    - deps: [mlflow]
+      framework: MLflow
+      type: mlops
+
+    - deps: [wandb]
+      framework: Weights & Biases
+      type: mlops
+
+  # Java 框架
+  java:
+    - deps: [spring-boot, org.springframework.boot]
+      framework: Spring Boot
+      type: backend
+
+    - deps: [spring-cloud]
+      framework: Spring Cloud
+      type: backend
+
+    - deps: [dubbo]
+      framework: Dubbo
+      type: backend
+
+    - deps: [quarkus]
+      framework: Quarkus
+      type: backend
+
+    - deps: [micronaut]
+      framework: Micronaut
+      type: backend
+
+    - deps: [vertx]
+      framework: Vert.x
+      type: backend
+
+  # Go 框架
+  go:
+    - deps: [gin-gonic]
+      framework: Gin
+      type: backend
+
+    - deps: [echo]
+      framework: Echo
+      type: backend
+
+    - deps: [fiber]
+      framework: Fiber
+      type: backend
+
+    - deps: [chi]
+      framework: Chi
+      type: backend
+
+  # Rust 框架
+  rust:
+    - deps: [actix]
+      framework: Actix
+      type: backend
+
+    - deps: [rocket]
+      framework: Rocket
+      type: backend
+
+    - deps: [warp]
+      framework: Warp
+      type: backend
+
+    - deps: [axum]
+      framework: Axum
+      type: backend
+
+# ============================================================
+# 数据开发组件检测规则
+# ============================================================
+dataComponents:
+  dataWarehouse:
+    label: 数据仓库
+    buildFiles: [dbt_project.yml]
+    directories: [models, marts, staging, seeds, snapshots]
+    filePatterns: ["*.sql", schema.yml, sources.yml]
+    type: data-warehouse
+    needsEvaluation: false
+
+  etlPipeline:
+    label: ETL管道
+    directories: [dags, pipelines, jobs, etl, workflows]
+    filePatterns: ["*dag*.py", "*pipeline*.py", "*etl*.py", "*job*.py"]
+    dependencies: [airflow, prefect, dagster, luigi, celery]
+    type: etl-pipeline
+    needsEvaluation: false
+
+  dataProcessing:
+    label: 数据处理
+    directories: [spark, processing, transform, scripts]
+    dependencies: [pyspark, dask, polars, duckdb, ray]
+    type: data-processing
+    needsEvaluation: false
+
+  dataQuality:
+    label: 数据质量
+    directories: [tests, data_tests, quality, expectations]
+    filePatterns: ["*test*.sql", "*quality*.py", great_expectations]
+    dependencies: [great_expectations, soda, dbt]
+    type: data-quality
+    needsEvaluation: false
+
+# ============================================================
+# AI 开发组件检测规则
+# ============================================================
+aiComponents:
+  modelTraining:
+    label: 模型训练
+    directories: [training, train, experiments, notebooks, models]
+    filePatterns: ["*train*.py", "*experiment*.py", "*.ipynb", train.sh]
+    dependencies: [torch, tensorflow, jax, keras, transformers, accelerate, deepspeed]
+    type: model-training
+    needsEvaluation: true
+
+  inferenceService:
+    label: 推理服务
+    directories: [inference, serving, api, endpoints]
+    filePatterns: ["*inference*.py", "*serve*.py", "*api*.py", serve.py]
+    dependencies: [fastapi, flask, uvicorn, tgi, vllm, triton, torchserve]
+    type: inference-service
+    needsEvaluation: true
+
+  llmApplication:
+    label: LLM应用
+    dependencies: [langchain, llamaindex, openai, anthropic, litellm, guidance, outlines]
+    type: llm-application
+    needsEvaluation: true
+
+  agentFramework:
+    label: Agent框架
+    dependencies: [langchain, crewai, autogen, metagpt, semantic-kernel, agents]
+    filePatterns: ["*agent*.py", "*agents*.py"]
+    type: agent-framework
+    needsEvaluation: true
+
+  vectorStore:
+    label: 向量存储
+    dependencies: [faiss, pinecone, weaviate, chromadb, qdrant, milvus, pgvector, lancedb]
+    type: vector-store
+    needsEvaluation: false
+
+  evaluation:
+    label: 评测框架
+    directories: [evaluation, eval, benchmarks, tests]
+    filePatterns: ["*eval*.py", "*benchmark*.py", "*metrics*.py"]
+    dependencies: [pytest, wandb, mlflow, evaluate, rouge, bert_score, sacrebleu]
+    type: evaluation
+    needsEvaluation: false
+
+  featureEngineering:
+    label: 特征工程
+    directories: [features, feature_store, feast]
+    dependencies: [feast, featuretools, fts]
+    type: feature-engineering
+    needsEvaluation: false
+
+  ragApplication:
+    label: RAG应用
+    requiresBoth: [vectorStore, llmApplication]
+    type: rag-application
+    needsEvaluation: true
+
+# ============================================================
+# 项目类型推断规则
+# ============================================================
+projectTypes:
+  monorepo:
+    indicators: [packages, apps, workspaces]
+    configFile: pnpm-workspace.yaml
+
+  microservices:
+    indicators: [services, backend, api]
+    minModules: 2
+
+  singleApp:
+    indicators: [src, lib, app]
+    excludeIndicators: [packages, services]
+
+# ============================================================
+# 入口文件模式
+# ============================================================
+entryPoints:
+  python: [main.py, app.py, manage.py, run.py, server.py, wsgi.py, asgi.py]
+  javascript: [index.js, app.js, server.js, main.js, index.ts, main.ts]
+  java: ["src/main/java/**/Application.java", "src/main/java/**/*Application.java"]
+  go: [main.go, "cmd/main.go", "cmd/server/main.go"]
+  rust: ["src/main.rs"]
+  ruby: [app.rb, config.ru]
+
+# ============================================================
+# 忽略目录（用于扫描）
+# ============================================================
+ignoreDirectories:
+  - node_modules
+  - ".git"
+  - ".svn"
+  - ".hg"
+  - dist
+  - build
+  - out
+  - target
+  - __pycache__
+  - ".pytest_cache"
+  - ".mypy_cache"
+  - venv
+  - env
+  - ".venv"
+  - ".env"
+  - ".idea"
+  - ".vscode"
+  - ".claude"
+  - ".autospec"
+  - data
+  - logs
+  - cache
+  - tmp
+  - temp
+
+# ============================================================
+# 源码目录模式
+# ============================================================
+sourceDirectories:
+  backend: [backend, server, api, service, services]
+  frontend: [frontend, web, client, ui]
+  mobile: [mobile, app]
+  data: [data, warehouse, datalake]
+  ai: [ml, ai, model, rag, agents]

package/knowledge/environment/middleware-knowledge.md

@@ -0,0 +1,237 @@
+# 中间件知识模板
+
+> 本模板定义中间件知识的标准格式。供 env-sync 生成知识文件时参考。
+
+---
+
+## 文件位置
+
+`.autospec/environment/env-knowledge/middleware/{name}.md`
+
+---
+
+## 基本信息
+
+| 字段 | 必填 | 说明 |
+|------|------|------|
+| 名称 | 是 | 中间件名称，如 mysql、redis |
+| 类型 | 是 | 参见类型枚举 |
+| 默认端口 | 否 | 默认端口号 |
+| 知识版本 | 是 | 格式 v1.0 |
+| 维护者 | 否 | 维护团队或个人 |
+
+### 类型枚举
+
+| 类型 | 说明 |
+|------|------|
+| relational-database | 关系型数据库 |
+| document-database | 文档数据库 |
+| cache | 缓存 |
+| message-queue | 消息队列 |
+| search-engine | 搜索引擎 |
+| vector-store | 向量存储 |
+| object-storage | 对象存储 |
+
+---
+
+## 环境变量
+
+### 必需变量
+
+| 变量名 | 说明 | 示例 |
+|--------|------|------|
+| {NAME}_HOST | 主机地址 | db.example.com |
+| {NAME}_DATABASE | 数据库名称 | myapp_db |
+| {NAME}_USER | 用户名 | app_user |
+| {NAME}_PASSWORD | 密码 | - |
+
+### 可选变量
+
+| 变量名 | 默认值 | 说明 |
+|--------|--------|------|
+| {NAME}_PORT | 默认端口 | 端口号 |
+| {NAME}_POOL_SIZE | 10 | 连接池大小 |
+| {NAME}_TIMEOUT | 5000 | 超时时间(ms) |
+
+---
+
+## 连接代码模板
+
+按语言分节，提供连接代码示例。
+
+### Node.js
+
+```javascript
+// 连接代码示例
+const client = new Client({
+  host: process.env.{NAME}_HOST,
+  port: process.env.{NAME}_PORT || {默认端口},
+  database: process.env.{NAME}_DATABASE,
+  user: process.env.{NAME}_USER,
+  password: process.env.{NAME}_PASSWORD,
+});
+```
+
+### Python
+
+```python
+# 连接代码示例
+import os
+client = Client(
+    host=os.environ['{NAME}_HOST'],
+    port=int(os.environ.get('{NAME}_PORT', {默认端口})),
+    database=os.environ['{NAME}_DATABASE'],
+    user=os.environ['{NAME}_USER'],
+    password=os.environ['{NAME}_PASSWORD'],
+)
+```
+
+### Java
+
+```java
+// 连接代码示例
+Client client = Client.builder()
+    .host(System.getenv("{NAME}_HOST"))
+    .port(Integer.parseInt(System.getenv().getOrDefault("{NAME}_PORT", "{默认端口}")))
+    .database(System.getenv("{NAME}_DATABASE"))
+    .user(System.getenv("{NAME}_USER"))
+    .password(System.getenv("{NAME}_PASSWORD"))
+    .build();
+```
+
+---
+
+## 最佳实践
+
+1. 使用连接池管理连接，避免每次请求创建新连接
+2. 设置合理的超时时间，避免连接泄漏
+3. 敏感信息使用环境变量，不要硬编码
+4. 实现重试机制，处理临时网络故障
+5. 记录连接日志，便于问题排查
+
+---
+
+## 反模式清单
+
+| 反模式 | 风险 | 替代方案 |
+|--------|------|----------|
+| 直接拼接SQL字符串 | SQL注入风险 | 使用参数化查询 |
+| 在代码中硬编码密码 | 安全风险 | 使用环境变量或密钥管理服务 |
+| 不设置超时时间 | 连接泄漏 | 设置合理的连接和查询超时 |
+| 每次请求创建新连接 | 性能问题 | 使用连接池 |
+| 忽略错误处理 | 系统不稳定 | 实现完善的错误处理和重试机制 |
+
+---
+
+## 健康检查
+
+```bash
+# 健康检查命令示例
+{CLI_TOOL} ping -h ${NAME}_HOST -u ${NAME}_USER -p${NAME}_PASSWORD
+```
+
+---
+
+## 示例：MySQL 中间件知识
+
+```markdown
+# MySQL 中间件知识
+
+## 基本信息
+
+| 字段 | 值 |
+|------|-----|
+| 名称 | mysql |
+| 类型 | relational-database |
+| 默认端口 | 3306 |
+| 知识版本 | v1.0 |
+
+## 环境变量
+
+### 必需变量
+
+| 变量名 | 说明 | 示例 |
+|--------|------|------|
+| MYSQL_HOST | 数据库主机地址 | db.example.com |
+| MYSQL_DATABASE | 数据库名称 | myapp_db |
+| MYSQL_USER | 用户名 | app_user |
+| MYSQL_PASSWORD | 密码 | - |
+
+### 可选变量
+
+| 变量名 | 默认值 | 说明 |
+|--------|--------|------|
+| MYSQL_PORT | 3306 | 数据库端口 |
+| MYSQL_POOL_SIZE | 10 | 连接池大小 |
+
+## 连接代码模板
+
+### Node.js
+
+```javascript
+const mysql = require('mysql2/promise');
+const pool = mysql.createPool({
+  host: process.env.MYSQL_HOST,
+  port: process.env.MYSQL_PORT || 3306,
+  database: process.env.MYSQL_DATABASE,
+  user: process.env.MYSQL_USER,
+  password: process.env.MYSQL_PASSWORD,
+  waitForConnections: true,
+  connectionLimit: parseInt(process.env.MYSQL_POOL_SIZE || '10')
+});
+```
+
+### Python
+
+```python
+import pymysql
+from dbutils.pooled_db import PooledDB
+
+pool = PooledDB(
+    creator=pymysql,
+    host=os.environ['MYSQL_HOST'],
+    port=int(os.environ.get('MYSQL_PORT', 3306)),
+    database=os.environ['MYSQL_DATABASE'],
+    user=os.environ['MYSQL_USER'],
+    password=os.environ['MYSQL_PASSWORD'],
+    maxconnections=int(os.environ.get('MYSQL_POOL_SIZE', '10'))
+)
+```
+
+## 最佳实践
+
+1. 使用连接池，不要在每次请求时创建新连接
+2. 设置合理的超时时间，避免连接泄漏
+3. 敏感信息使用环境变量，不要硬编码
+
+## 反模式清单
+
+| 反模式 | 风险 | 替代方案 |
+|--------|------|----------|
+| 直接拼接SQL字符串 | SQL注入风险 | 使用参数化查询 |
+| 在代码中硬编码密码 | 安全风险 | 使用环境变量 |
+
+## 健康检查
+
+```bash
+mysqladmin ping -h ${MYSQL_HOST} -u ${MYSQL_USER} -p${MYSQL_PASSWORD}
+```
+```
+
+---
+
+## 自检清单
+
+生成中间件知识文件后，检查以下项目：
+
+- [ ] 基本信息章节完整，包含名称、类型、知识版本
+- [ ] 环境变量章节包含必需变量和可选变量表格
+- [ ] 连接代码模板包含项目使用的主要语言
+- [ ] 最佳实践列表不为空
+- [ ] 反模式清单包含风险和替代方案
+
+---
+
+**维护者**: AutoSpec 团队
+**进化分区**: 🟡 受控区
+**关联文档**: `knowledge/environment/template-registry.md`