langchain4j-testing-strategies

Name: langchain4j-testing-strategies
Author: giuseppe-trisciuoglio

1.6kinstalls36forks

How to install langchain4j-testing-strategies?

giuseppe-trisciuoglio/developer-kit/langchain4j-testing-strategies

$npx -y skills add giuseppe-trisciuoglio/developer-kit --skill langchain4j-testing-strategies

Installs into the current project.

Summary

Provides unit test, integration test, and mock AI patterns for LangChain4j applications. Creates mock LLM responses, tests retrieval chains, validates RAG workflows, and implements Testcontainers-based integration tests for Java AI services. Use when unit testing AI services, int

Files · 1

View on GitHub

SKILL.md

1	# LangChain4J Testing Strategies
2
3	## Overview
4
5	Patterns for unit testing with mocks, integration testing with Testcontainers, and end-to-end validation of RAG systems, AI Services, and tool execution.
6
7	## When to Use
8
9	- Unit testing AI services: When you need fast, isolated tests for services using LangChain4j AiServices
10	- Integration testing LangChain4j components: When testing real ChatModel, EmbeddingModel, or RAG pipelines with Testcontainers
11	- Mocking AI models: When you need deterministic responses without calling external APIs
12	- Testing LLM-based Java applications: When validating RAG workflows, tool execution, or retrieval chains
13
14	## Instructions
15
16	### 1. Unit Testing with Mocks
17
18	Use mock models for fast, isolated testing. See `references/unit-testing.md`.
19
20	```java
21	ChatModel mockModel = mock(ChatModel.class);
22	when(mockModel.generate(any(String.class)))
23	.thenReturn(Response.from(AiMessage.from("Mocked response")));
24
25	var service = AiServices.builder(AiService.class)
26	.chatModel(mockModel)
27	.build();
28	```
29
30	### 2. Configure Testing Dependencies
31
32	Setup Maven/Gradle dependencies. See `references/testing-dependencies.md`.
33
34	- `langchain4j-test` - Guardrail assertions
35	- `testcontainers` - Containerized testing
36	- `mockito` - Mock external dependencies
37	- `assertj` - Fluent assertions
38
39	### 3. Integration Testing with Testcontainers
40
41	Test with real services. See `references/integration-testing.md`.
42
43	```java
44	@Testcontainers
45	class OllamaIntegrationTest {
46	@Container
47	static GenericContainer<?> ollama = new GenericContainer<>(
48	DockerImageName.parse("ollama/ollama:0.5.4")
49	).withExposedPorts(11434);
50
51	@Test
52	void shouldGenerateResponse() {
53	// Verify container is healthy
54	assertTrue(ollama.isRunning());
55	await().atMost(30, TimeUnit.SECONDS)
56	.until(() -> ollama.getLogs().contains("API server listening"));
57
58	ChatModel model = OllamaChatModel.builder()
59	.baseUrl(ollama.getEndpoint())
60	.build();
61
62	// Verify model responds before running tests
63	assertDoesNotThrow(() -> model.generate("ping"));
64
65	String response = model.generate("Test query");
66	assertNotNull(response);
67	}
68	}
69	```
70
71	### 4. Advanced Features
72
73	Streaming, memory, error handling patterns in `references/advanced-testing.md`.
74
75	### 5. Testing Workflow
76
77	Follow the testing pyramid from `references/workflow-patterns.md`:
78
79	- 70% Unit Tests: Fast, isolated with mocks
80	- 20% Integration Tests: Real services with health checks
81	- 10% End-to-End Tests: Complete workflows
82
83	```
84	70% Unit Tests ─ Mock ChatModel, guardrails, edge cases
85	20% Integration Tests ─ Testcontainers, vector stores, RAG
86	10% End-to-End Tests ─ Complete user journeys
87	```
88
89	### Troubleshooting
90
91	- Container fails to start: Check Docker daemon is running, verify image exists, increase timeout
92	- Model not responding: Verify baseUrl is correct, check container logs, ensure model is loaded
93	- Test timeout: Increase `@Timeout` duration for slow models, check container resource limits
94	- Flaky tests: Add retry logic or health checks before assertions
95
96	## Examples
97
98	### Unit Test
99
100	```java
101	@Test
102	void shouldProcessQueryWithMock() {
103	ChatModel mockModel = mock(ChatModel.class);
104	when(mockModel.generate(any(String.class)))
105	.thenReturn(Response.from(AiMessage.from("Test response")));
106
107	var service = AiServices.builder(AiService.class)
108	.chatModel(mockModel)
109	.build();
110
111	String result = service.chat("What is Java?");
112	assertEquals("Test response", result);
113	}
114	```
115
116	### Integration Test with Testcontainers
117
118	```java
119	@Testcontainers
120	class RAGIntegrationTest {
121	@Container
122	static GenericContainer<?> ollama = new GenericContainer<>(
123	DockerImageName.parse("ollama/ollama:0.5.4")
124	);
125
126	@BeforeAll
127	static void waitForContainerReady() {
128	await().atMost(60, TimeUnit.SECONDS)
129	.until(() -> ollama.getLogs().contains("API server listening"));
130	}
131
132	@Test
133	void shouldCompleteRAGWorkflow() {
134	assertTrue(ollama.isRunning());
135
136	var chatModel = OllamaChatModel.builder()
137	.baseUrl(ollama.getEndpoint())
138	.build();
139
140	var embeddingModel = OllamaEmbeddingModel.builder()
141	.baseUrl(ollama.getEndpoint())
142	.build();
143
144	var store = new InMemoryEmbeddingStore<>();
145	var retriever = EmbeddingStoreContentRetriever.bu

People who liked this also viewed

6 picks

Type

langchain4j-testing-strategies

1.6kinstalls36forks

View on GitHub

How to install langchain4j-testing-strategies?

giuseppe-trisciuoglio/developer-kit/langchain4j-testing-strategies

$npx -y skills add giuseppe-trisciuoglio/developer-kit --skill langchain4j-testing-strategies

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# LangChain4J Testing Strategies
2
3	## Overview
4
5	Patterns for unit testing with mocks, integration testing with Testcontainers, and end-to-end validation of RAG systems, AI Services, and tool execution.
6
7	## When to Use
8
9	- Unit testing AI services: When you need fast, isolated tests for services using LangChain4j AiServices
10	- Integration testing LangChain4j components: When testing real ChatModel, EmbeddingModel, or RAG pipelines with Testcontainers
11	- Mocking AI models: When you need deterministic responses without calling external APIs
12	- Testing LLM-based Java applications: When validating RAG workflows, tool execution, or retrieval chains
13
14	## Instructions
15
16	### 1. Unit Testing with Mocks
17
18	Use mock models for fast, isolated testing. See `references/unit-testing.md`.
19
20	```java
21	ChatModel mockModel = mock(ChatModel.class);
22	when(mockModel.generate(any(String.class)))
23	.thenReturn(Response.from(AiMessage.from("Mocked response")));
24
25	var service = AiServices.builder(AiService.class)
26	.chatModel(mockModel)
27	.build();
28	```
29
30	### 2. Configure Testing Dependencies
31
32	Setup Maven/Gradle dependencies. See `references/testing-dependencies.md`.
33
34	- `langchain4j-test` - Guardrail assertions
35	- `testcontainers` - Containerized testing
36	- `mockito` - Mock external dependencies
37	- `assertj` - Fluent assertions
38
39	### 3. Integration Testing with Testcontainers
40
41	Test with real services. See `references/integration-testing.md`.
42
43	```java
44	@Testcontainers
45	class OllamaIntegrationTest {
46	@Container
47	static GenericContainer<?> ollama = new GenericContainer<>(
48	DockerImageName.parse("ollama/ollama:0.5.4")
49	).withExposedPorts(11434);
50
51	@Test
52	void shouldGenerateResponse() {
53	// Verify container is healthy
54	assertTrue(ollama.isRunning());
55	await().atMost(30, TimeUnit.SECONDS)
56	.until(() -> ollama.getLogs().contains("API server listening"));
57
58	ChatModel model = OllamaChatModel.builder()
59	.baseUrl(ollama.getEndpoint())
60	.build();
61
62	// Verify model responds before running tests
63	assertDoesNotThrow(() -> model.generate("ping"));
64
65	String response = model.generate("Test query");
66	assertNotNull(response);
67	}
68	}
69	```
70
71	### 4. Advanced Features
72
73	Streaming, memory, error handling patterns in `references/advanced-testing.md`.
74
75	### 5. Testing Workflow
76
77	Follow the testing pyramid from `references/workflow-patterns.md`:
78
79	- 70% Unit Tests: Fast, isolated with mocks
80	- 20% Integration Tests: Real services with health checks
81	- 10% End-to-End Tests: Complete workflows
82
83	```
84	70% Unit Tests ─ Mock ChatModel, guardrails, edge cases
85	20% Integration Tests ─ Testcontainers, vector stores, RAG
86	10% End-to-End Tests ─ Complete user journeys
87	```
88
89	### Troubleshooting
90
91	- Container fails to start: Check Docker daemon is running, verify image exists, increase timeout
92	- Model not responding: Verify baseUrl is correct, check container logs, ensure model is loaded
93	- Test timeout: Increase `@Timeout` duration for slow models, check container resource limits
94	- Flaky tests: Add retry logic or health checks before assertions
95
96	## Examples
97
98	### Unit Test
99
100	```java
101	@Test
102	void shouldProcessQueryWithMock() {
103	ChatModel mockModel = mock(ChatModel.class);
104	when(mockModel.generate(any(String.class)))
105	.thenReturn(Response.from(AiMessage.from("Test response")));
106
107	var service = AiServices.builder(AiService.class)
108	.chatModel(mockModel)
109	.build();
110
111	String result = service.chat("What is Java?");
112	assertEquals("Test response", result);
113	}
114	```
115
116	### Integration Test with Testcontainers
117
118	```java
119	@Testcontainers
120	class RAGIntegrationTest {
121	@Container
122	static GenericContainer<?> ollama = new GenericContainer<>(
123	DockerImageName.parse("ollama/ollama:0.5.4")
124	);
125
126	@BeforeAll
127	static void waitForContainerReady() {
128	await().atMost(60, TimeUnit.SECONDS)
129	.until(() -> ollama.getLogs().contains("API server listening"));
130	}
131
132	@Test
133	void shouldCompleteRAGWorkflow() {
134	assertTrue(ollama.isRunning());
135
136	var chatModel = OllamaChatModel.builder()
137	.baseUrl(ollama.getEndpoint())
138	.build();
139
140	var embeddingModel = OllamaEmbeddingModel.builder()
141	.baseUrl(ollama.getEndpoint())
142	.build();
143
144	var store = new InMemoryEmbeddingStore<>();
145	var retriever = EmbeddingStoreContentRetriever.bu

People who liked this also viewed

6 picks

Type