Contexa Examples

Contexa AI Native Zero Trust Security Platform의 실전 예제 모음. 각 모듈은 독립적인 Spring Boot 애플리케이션으로, 가장 단순한 설정부터 고급 구성까지 단계적으로 학습할 수 있다.

Prerequisites

Requirement	Version
Java	21+
PostgreSQL	15+ (pgvector extension)
Ollama	latest
Gradle	8.x (Wrapper included)

Database Setup

CREATE DATABASE contexa;
\c contexa
CREATE EXTENSION IF NOT EXISTS vector;

Ollama Models

ollama pull qwen2.5:7b
ollama pull mxbai-embed-large

Dependency

All modules depend on spring-boot-starter-contexa from Maven Local:

implementation 'ai.ctxa:spring-boot-starter-contexa:0.1.0'

---

Modules

Get Started

Module	Port	Description
contexa-example-quickstart	8081	Minimal setup: @EnableAISecurity + YAML
contexa-example-protectable	8082	@Protectable method-level AI security
contexa-example-identity-dsl	8083	Identity DSL: custom auth flows
contexa-example-shadow-enforce	8084	Shadow/Enforce mode transition

Identity

Module	Port	Description
contexa-example-identity-rest	8085	REST API authentication (JSON, SPA/Mobile)
contexa-example-identity-ott	8086	One-Time Token / Magic Link
contexa-example-identity-mfa	8087	Advanced MFA with custom pages
contexa-example-identity-asep	8088	ASEP security exception annotations
contexa-example-identity-oauth2	8089	OAuth2 JWT stateless authentication

IAM

Module	Port	Description
contexa-example-iam-dynamic-auth	8090	URL dynamic authorization with AI security expressions (#trust, #ai)
contexa-example-iam-protectable-analysis	8091	@Protectable + 5 AnalysisRequirement levels + SSE real-time LLM analysis
contexa-example-iam-permission	8092	hasPermission() with custom DomainPermissionEvaluator

AI Engine

Module	Port	Description
contexa-example-ai-lab	8093	Custom AILab + AIStrategy + PromptTemplate (sentiment analysis)
contexa-example-ai-pipeline	8094	PipelineOrchestrator + Custom PipelineStep + DomainResponseProcessor + SSE streaming
contexa-example-ai-llm	8095	UnifiedLLMOrchestrator + Custom BaseAdvisor + ExecutionContext + Chat UI

---

Running Examples

Each module runs independently:

# Quickstart (port 8081)
./gradlew :contexa-example-quickstart:bootRun

# Protectable (port 8082)
./gradlew :contexa-example-protectable:bootRun

# Identity DSL (port 8083)
./gradlew :contexa-example-identity-dsl:bootRun

# Shadow Mode (port 8084)
./gradlew :contexa-example-shadow-enforce:bootRun --args='--spring.profiles.active=shadow'

# Enforce Mode (port 8084)
./gradlew :contexa-example-shadow-enforce:bootRun --args='--spring.profiles.active=enforce'

# Identity REST (port 8085)
./gradlew :contexa-example-identity-rest:bootRun

# Identity OTT (port 8086)
./gradlew :contexa-example-identity-ott:bootRun

# Identity MFA (port 8087)
./gradlew :contexa-example-identity-mfa:bootRun

# Identity ASEP (port 8088)
./gradlew :contexa-example-identity-asep:bootRun

# Identity OAuth2 (port 8089)
./gradlew :contexa-example-identity-oauth2:bootRun

# IAM Dynamic Auth (port 8090)
./gradlew :contexa-example-iam-dynamic-auth:bootRun

# IAM Protectable Analysis (port 8091)
./gradlew :contexa-example-iam-protectable-analysis:bootRun

# IAM Permission (port 8092)
./gradlew :contexa-example-iam-permission:bootRun

# AI Lab (port 8093) — http://localhost:8093/lab/test
./gradlew :contexa-example-ai-lab:bootRun

# AI Pipeline (port 8094) — http://localhost:8094/pipeline/test
./gradlew :contexa-example-ai-pipeline:bootRun

# AI LLM (port 8095) — http://localhost:8095/llm/test
./gradlew :contexa-example-ai-llm:bootRun

---

Common Configuration

All modules share the following base configuration:

contexa:
  infrastructure:
    mode: standalone

spring:
  datasource:
    url: jdbc:postgresql://localhost:5432/contexa
    username: ${DB_USERNAME:admin}
    password: ${DB_PASSWORD:1111}
    driver-class-name: org.postgresql.Driver
  jpa:
    database: POSTGRESQL
    hibernate:
      ddl-auto: update
  ai:
    chat:
      model:
        priority: ollama
    ollama:
      base-url: http://127.0.0.1:11434
      chat:
        options:
          model: qwen2.5:7b
          keep-alive: "24h"
      embedding:
        enabled: true
        model: mxbai-embed-large
    vectorstore:
      pgvector:
        table-name: vector_store
        index-type: HNSW
        distance-type: COSINE_DISTANCE
        dimensions: 1024
        initialize-schema: true

Environment variables DB_USERNAME and DB_PASSWORD can override database credentials.

---

Build

# Compile all modules
./gradlew compileJava

# Build all modules
./gradlew build