java-spring-ai

Name: java-spring-ai
Author: ducpm2303

17stars0forks

How to install java-spring-ai?

ducpm2303/claude-java-plugins/java-spring-ai

$npx -y skills add ducpm2303/claude-java-plugins --skill java-spring-ai

Installs into the current project.

Summary

Use when the user asks to add AI features, integrate Spring AI or LangChain4J, build a chatbot, implement RAG (retrieval-augmented generation), use vector stores, stream LLM responses, or call AI tools/functions in a Spring Boot project.

Files · 1

View on GitHub

SKILL.md

1	# Spring AI / LangChain4J Skill
2
3	Detect the framework in use, then apply the correct patterns.
4
5	## Step 1 — Detect framework and version
6
7	Check `pom.xml` or `build.gradle`:
8	- `spring-ai-` dependency → Spring AI* (note version: 1.0.x GA or 0.8.x milestone)
9	- `langchain4j-` dependency → LangChain4J* (note version: 0.x or 1.x)
10	- Neither present → offer to add one (recommend Spring AI for Spring Boot 3.x, LangChain4J for Boot 2.x)
11
12	Check Spring Boot version:
13	- Boot 3.x → Spring AI 1.x preferred, LangChain4J 0.35+
14	- Boot 2.x → LangChain4J 0.30.x (Spring AI requires Boot 3.x)
15
16	---
17
18	## Mode: `review`
19
20	User asks to review existing AI code. Check for:
21
22	Spring AI:
23	- [ ] `ChatClient` built via `ChatClient.Builder` (not raw `ChatModel`) for fluent API
24	- [ ] Prompt templates use `PromptTemplate` with variables — no string concatenation
25	- [ ] Streaming uses `stream().content()` or `Flux<String>` — not blocking `.call()` for real-time responses
26	- [ ] `@Retryable` or Spring AI retry config on ChatClient calls — LLMs are flaky
27	- [ ] Secrets (`spring.ai.openai.api-key`) come from env vars or Vault, never hardcoded
28	- [ ] `VectorStore` queries use `SearchRequest.query(text).withTopK(n)` — not raw SQL
29	- [ ] RAG advisor (`QuestionAnswerAdvisor`) attached to ChatClient — not manual context injection
30	- [ ] Token usage logged at DEBUG, not INFO (avoid log noise)
31
32	LangChain4J:
33	- [ ] AI services use `@AiService` interface — not `ChatLanguageModel.generate()` directly
34	- [ ] System prompts in `@SystemMessage` annotation — not hardcoded strings
35	- [ ] Memory uses `MessageWindowChatMemory` or `TokenWindowChatMemory` — not unlimited history
36	- [ ] Streaming via `StreamingChatLanguageModel` with `TokenStream` — not blocking
37	- [ ] Embeddings via `EmbeddingModel` + `EmbeddingStore` for RAG — not in-memory list search
38	- [ ] Tools annotated with `@Tool` on service methods — not manual function dispatch
39	- [ ] API key from `@Value("${langchain4j.openai.api-key}")` — never literal
40
41	---
42
43	## Mode: `chat`
44
45	User asks to add a basic chatbot or chat endpoint.
46
47	### Spring AI
48	1. Add dependency (see `references/patterns.md` → Spring AI Setup)
49	2. Inject `ChatClient.Builder`, build a `ChatClient` bean
50	3. Create `ChatController` with `@PostMapping("/chat")`
51	4. Use `chatClient.prompt().user(message).call().content()` for simple response
52	5. For streaming: return `Flux<String>` with `chatClient.prompt().user(message).stream().content()`
53	6. Add `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` to `application.yml` via `${env-var}`
54
55	### LangChain4J
56	1. Add `langchain4j-spring-boot-starter` + provider dependency
57	2. Define `@AiService` interface with `@SystemMessage`
58	3. Register as Spring bean via `AiServices.builder(MyAssistant.class).chatLanguageModel(model).build()`
59	4. Expose via `@RestController`
60
61	---
62
63	## Mode: `rag`
64
65	User asks to implement RAG (chat over documents, knowledge base, semantic search).
66
67	### Spring AI RAG
68	1. Choose vector store: PgVector (PostgreSQL), Chroma, Redis, Weaviate, Qdrant (see `references/patterns.md`)
69	2. Add `spring-ai-{store}-store-spring-boot-starter`
70	3. Ingest pipeline:
71	- `DocumentReader` (PDF, text, web) → `TokenTextSplitter` → `VectorStore.add()`
72	- Run at startup via `ApplicationRunner` or dedicated `@PostMapping("/ingest")`
73	4. Query pipeline:
74	- Attach `QuestionAnswerAdvisor(vectorStore)` to `ChatClient`
75	- Spring AI auto-retrieves context and injects into prompt
76	5. Tune: `SearchRequest.withTopK(5).withSimilarityThreshold(0.7)`
77
78	### LangChain4J RAG
79	1. Add `EmbeddingStore` (Chroma, Qdrant, in-memory for dev)
80	2. `EmbeddingStoreIngestor` with `DocumentSplitter` and `EmbeddingModel`
81	3. `EmbeddingStoreContentRetriever` → `RetrievalAugmentor` → `AiServices` builder
82
83	---
84
85	## Mode: `tools`
86
87	User asks to give the AI the ability to call Java methods (function/tool calling).
88
89	### Spring AI
90	1. Define a `@Bean` of type `Function<Input, Output>` — Spring AI auto-registers it
91	2. Or use `@Description` on a `record` parameter for rich schema
92	3. Pass function names to `ChatClient`: `.options(OpenAiChatOptions.builder().withFunction("myFunction").build())`
93	4. Spring AI handles the tool call loop automatically
94
95	### LangChain4J
96	1. Annotate service methods with `@Tool("description of what this tool does")`
97	2. Register the service as a tool: `AiServices.builder(...).tools(myToolService).build()`
98	3. The model decides when to call — no manual dispatch needed
99
100	---
101
102	## Mode: `memory`
103
104	User asks to add conversation memory / chat history.
105
106	### Spring AI
107	- `MessageChatMemoryAdvisor` with `InMemoryCh

People who liked this also viewed

6 picks

Type

java-spring-ai

17stars0forks

View on GitHub

How to install java-spring-ai?

ducpm2303/claude-java-plugins/java-spring-ai

$npx -y skills add ducpm2303/claude-java-plugins --skill java-spring-ai

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Spring AI / LangChain4J Skill
2
3	Detect the framework in use, then apply the correct patterns.
4
5	## Step 1 — Detect framework and version
6
7	Check `pom.xml` or `build.gradle`:
8	- `spring-ai-` dependency → Spring AI* (note version: 1.0.x GA or 0.8.x milestone)
9	- `langchain4j-` dependency → LangChain4J* (note version: 0.x or 1.x)
10	- Neither present → offer to add one (recommend Spring AI for Spring Boot 3.x, LangChain4J for Boot 2.x)
11
12	Check Spring Boot version:
13	- Boot 3.x → Spring AI 1.x preferred, LangChain4J 0.35+
14	- Boot 2.x → LangChain4J 0.30.x (Spring AI requires Boot 3.x)
15
16	---
17
18	## Mode: `review`
19
20	User asks to review existing AI code. Check for:
21
22	Spring AI:
23	- [ ] `ChatClient` built via `ChatClient.Builder` (not raw `ChatModel`) for fluent API
24	- [ ] Prompt templates use `PromptTemplate` with variables — no string concatenation
25	- [ ] Streaming uses `stream().content()` or `Flux<String>` — not blocking `.call()` for real-time responses
26	- [ ] `@Retryable` or Spring AI retry config on ChatClient calls — LLMs are flaky
27	- [ ] Secrets (`spring.ai.openai.api-key`) come from env vars or Vault, never hardcoded
28	- [ ] `VectorStore` queries use `SearchRequest.query(text).withTopK(n)` — not raw SQL
29	- [ ] RAG advisor (`QuestionAnswerAdvisor`) attached to ChatClient — not manual context injection
30	- [ ] Token usage logged at DEBUG, not INFO (avoid log noise)
31
32	LangChain4J:
33	- [ ] AI services use `@AiService` interface — not `ChatLanguageModel.generate()` directly
34	- [ ] System prompts in `@SystemMessage` annotation — not hardcoded strings
35	- [ ] Memory uses `MessageWindowChatMemory` or `TokenWindowChatMemory` — not unlimited history
36	- [ ] Streaming via `StreamingChatLanguageModel` with `TokenStream` — not blocking
37	- [ ] Embeddings via `EmbeddingModel` + `EmbeddingStore` for RAG — not in-memory list search
38	- [ ] Tools annotated with `@Tool` on service methods — not manual function dispatch
39	- [ ] API key from `@Value("${langchain4j.openai.api-key}")` — never literal
40
41	---
42
43	## Mode: `chat`
44
45	User asks to add a basic chatbot or chat endpoint.
46
47	### Spring AI
48	1. Add dependency (see `references/patterns.md` → Spring AI Setup)
49	2. Inject `ChatClient.Builder`, build a `ChatClient` bean
50	3. Create `ChatController` with `@PostMapping("/chat")`
51	4. Use `chatClient.prompt().user(message).call().content()` for simple response
52	5. For streaming: return `Flux<String>` with `chatClient.prompt().user(message).stream().content()`
53	6. Add `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` to `application.yml` via `${env-var}`
54
55	### LangChain4J
56	1. Add `langchain4j-spring-boot-starter` + provider dependency
57	2. Define `@AiService` interface with `@SystemMessage`
58	3. Register as Spring bean via `AiServices.builder(MyAssistant.class).chatLanguageModel(model).build()`
59	4. Expose via `@RestController`
60
61	---
62
63	## Mode: `rag`
64
65	User asks to implement RAG (chat over documents, knowledge base, semantic search).
66
67	### Spring AI RAG
68	1. Choose vector store: PgVector (PostgreSQL), Chroma, Redis, Weaviate, Qdrant (see `references/patterns.md`)
69	2. Add `spring-ai-{store}-store-spring-boot-starter`
70	3. Ingest pipeline:
71	- `DocumentReader` (PDF, text, web) → `TokenTextSplitter` → `VectorStore.add()`
72	- Run at startup via `ApplicationRunner` or dedicated `@PostMapping("/ingest")`
73	4. Query pipeline:
74	- Attach `QuestionAnswerAdvisor(vectorStore)` to `ChatClient`
75	- Spring AI auto-retrieves context and injects into prompt
76	5. Tune: `SearchRequest.withTopK(5).withSimilarityThreshold(0.7)`
77
78	### LangChain4J RAG
79	1. Add `EmbeddingStore` (Chroma, Qdrant, in-memory for dev)
80	2. `EmbeddingStoreIngestor` with `DocumentSplitter` and `EmbeddingModel`
81	3. `EmbeddingStoreContentRetriever` → `RetrievalAugmentor` → `AiServices` builder
82
83	---
84
85	## Mode: `tools`
86
87	User asks to give the AI the ability to call Java methods (function/tool calling).
88
89	### Spring AI
90	1. Define a `@Bean` of type `Function<Input, Output>` — Spring AI auto-registers it
91	2. Or use `@Description` on a `record` parameter for rich schema
92	3. Pass function names to `ChatClient`: `.options(OpenAiChatOptions.builder().withFunction("myFunction").build())`
93	4. Spring AI handles the tool call loop automatically
94
95	### LangChain4J
96	1. Annotate service methods with `@Tool("description of what this tool does")`
97	2. Register the service as a tool: `AiServices.builder(...).tools(myToolService).build()`
98	3. The model decides when to call — no manual dispatch needed
99
100	---
101
102	## Mode: `memory`
103
104	User asks to add conversation memory / chat history.
105
106	### Spring AI
107	- `MessageChatMemoryAdvisor` with `InMemoryCh

People who liked this also viewed

6 picks

Type