Qdrant Agent Skill

Provides Qdrant vector database integration patterns with LangChain4j. Handles embedding storage, similarity search, and vector management for Java applications. Use when implementing vector-based retrieval for RAG systems, semantic search, or recommendation engines.

7k tokens

context cost

the whole folder, loaded on every use

files

instructions only

copies elsewhere

how many repositories repackaged it

316

stars on the repo

on the repository, not the skill itself

Install

one command, takes just this skill from the repository

npx skills add https://github.com/giuseppe-trisciuoglio/developer-kit --skill qdrant

The instruction itself

20 sections, as written by the author

Qdrant Vector Database Integration

Overview

Qdrant is an AI-native vector database for semantic search and similarity retrieval. This skill provides patterns for integrating Qdrant with Java applications, focusing on Spring Boot and LangChain4j integration.

When to Use

Semantic search or recommendation systems in Spring Boot applications
RAG pipelines with Java and LangChain4j
Vector database integration for AI/ML applications
High-performance similarity search with filtered queries

Instructions

1. Deploy Qdrant with Docker

docker run -p 6333:6333 -p 6334:6334 \
    -v "$(pwd)/qdrant_storage:/qdrant/storage:z" \
    qdrant/qdrant

Access: REST API at http://localhost:6333, gRPC at http://localhost:6334.

2. Add Dependencies

Maven:

<dependency>
    <groupId>io.qdrant</groupId>
    <artifactId>client</artifactId>
    <version>1.15.0</version>
</dependency>

Gradle:

implementation 'io.qdrant:client:1.15.0'

3. Initialize Client

QdrantClient client = new QdrantClient(
    QdrantGrpcClient.newBuilder("localhost").build());

For production with API key:

QdrantClient client = new QdrantClient(
    QdrantGrpcClient.newBuilder("localhost", 6334, false)
        .withApiKey("YOUR_API_KEY")
        .build());

4. Create Collection

client.createCollectionAsync("search-collection",
    VectorParams.newBuilder()
        .setDistance(Distance.Cosine)
        .setSize(384)
        .build()
).get();

Validation: Verify the collection was created by checking client.getCollectionAsync("search-collection").get().

5. Upsert Vectors

List<PointStruct> points = List.of(
    PointStruct.newBuilder()
        .setId(id(1))
        .setVectors(vectors(0.05f, 0.61f, 0.76f, 0.74f))
        .putAllPayload(Map.of("title", value("Spring Boot Documentation")))
        .build()
);
client.upsertAsync("search-collection", points).get();

Validation: Check that client.upsertAsync(...).get() completes without throwing.

6. Search Vectors

List<ScoredPoint> results = client.queryAsync(
    QueryPoints.newBuilder()
        .setCollectionName("search-collection")
        .setLimit(5)
        .setQuery(nearest(0.2f, 0.1f, 0.9f, 0.7f))
        .build()
).get();

Filtered search:

List<ScoredPoint> results = client.searchAsync(
    SearchPoints.newBuilder()
        .setCollectionName("search-collection")
        .addAllVector(List.of(0.62f, 0.12f, 0.53f, 0.12f))
        .setFilter(Filter.newBuilder()
            .addMust(range("category", Range.newBuilder().setEq("docs").build()))
            .build())
        .setLimit(5)
        .build()).get();

LangChain4j Integration

For RAG pipelines, use LangChain4j's high-level abstractions:

EmbeddingStore<TextSegment> embeddingStore = QdrantEmbeddingStore.builder()
    .collectionName("rag-collection")
    .host("localhost")
    .port(6334)
    .apiKey("YOUR_API_KEY")
    .build();

Spring Boot configuration with LangChain4j:

@Bean
public EmbeddingStore<TextSegment> embeddingStore() {
    return QdrantEmbeddingStore.builder()
        .collectionName("rag-collection")
        .host(host)
        .port(port)
        .build();
}

@Bean
public EmbeddingModel embeddingModel() {
    return new AllMiniLmL6V2EmbeddingModel();
}

Spring Boot Integration

Inject the client via configuration:

@Configuration
public class QdrantConfig {
    @Value("${qdrant.host:localhost}")
    private String host;

    @Value("${qdrant.port:6334}")
    private int port;

    @Bean
    public QdrantClient qdrantClient() {
        return new QdrantClient(
            QdrantGrpcClient.newBuilder(host, port, false).build());
    }
}

Examples

REST Search Endpoint

@RestController
@RequestMapping("/api/search")
public class SearchController {
    private final VectorSearchService searchService;

    public SearchController(VectorSearchService searchService) {
        this.searchService = searchService;
    }

    @GetMapping
    public List<ScoredPoint> search(@RequestParam String query) {
        List<Float> queryVector = embeddingModel.embed(query).content().vectorAsList();
        return searchService.search("documents", queryVector);
    }
}

Best Practices

Distance metric: Cosine for normalized text embeddings, Euclidean for non-normalized.
Batch upserts: Use batch operations over individual point insertions.
Connection pooling: Configure connection pooling for high-throughput production workloads.
Error handling: Wrap async operations in try/catch for ExecutionException/InterruptedException.
API keys: Store in environment variables or Spring config, never hardcode.

Advanced Patterns

Multi-tenant Storage

public void upsertForTenant(String tenantId, List<PointStruct> points) {
    String collectionName = "tenant_" + tenantId + "_documents";
    client.upsertAsync(collectionName, points).get();
}

Docker Compose for Production

services:
  qdrant:
    image: qdrant/qdrant:v1.7.0
    ports:
      - "6333:6333"
      - "6334:6334"
    volumes:
      - qdrant_storage:/qdrant/storage

References

Qdrant API Reference — Complete client API documentation
Complete Spring Boot Examples — Full application implementations
Qdrant Documentation
LangChain4j Documentation

Constraints and Warnings

Vector dimensions must match the embedding model exactly; mismatched dimensions cause upsert errors.
Input validation: Sanitize all document content before ingestion; untrusted payloads may contain prompt injection attacks.
Content filtering: Apply content filtering on retrieved documents before passing them to the LLM.
Large collections require proper indexing for acceptable search performance.
Use gRPC API (port 6334) for production; REST API (port 6333) for debugging only.
Collection recreation deletes all data; implement backup strategies for production environments.

Other skills for the same job

different authors, same section of the catalogue

Skill Creator

by anthropics

vendor ×10

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Access NCBI GEO for gene expression/genomics data. Search/download microarray and RNA-seq datasets (GSE, GSM, GPL), retrieve SOFT/Matrix files, for transcriptomics and expression analysis.

12k tokens

Pymc Bayesian Modeling

by christophacham

×4

Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming and inference.

Multi-objective optimization framework. NSGA-II, NSGA-III, MOEA/D, Pareto fronts, constraint handling, benchmarks (ZDT, DTLZ), for engineering design and optimization problems.

Statistical modeling toolkit. OLS, GLM, logistic, ARIMA, time series, hypothesis tests, diagnostics, AIC/BIC, for rigorous statistical inference and econometric analysis.

Add unsigned integer (uint) type support to PyTorch operators by updating AT_DISPATCH macros. Use when adding support for uint16, uint32, uint64 types to operators, kernels, or when user mentions enabling unsigned types, barebones unsigned types, or uint support.

Convert PyTorch AT_DISPATCH macros to AT_DISPATCH_V2 format in ATen C++ code. Use when porting AT_DISPATCH_ALL_TYPES_AND*, AT_DISPATCH_FLOATING_TYPES*, or other dispatch macros to the new v2 API. For ATen kernel files, CUDA kernels, and native operator implementations.

Write docstrings for PyTorch functions and methods following PyTorch conventions. Use when writing or updating docstrings in PyTorch code.

3k tokens

How to use it

Copy the folder

Take giuseppe-trisciuoglio/qdrant from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.

Install what it needs

The instructions reference docker. Without those the skill loads but fails at the first command.